Crysis On New Mac Pro With 8800!!!

[ildondeigiocchi]

I am still awaiting my Mac Pro to arrive with the 8800GT. I can't wait to play Crysis on it. To all people who received the 8800GT Mac Pro, can you please post results of gameplay and specs you were palying at OR POST A VIDEO OF THE GAMEPLAy on the Mac Pro with 8800. Please post a video using fraps. CRYSIS RULES!!!! :):apple::)

[Eidorian]

You could just look at every 8800GT Crysis benchmark and review that's online. I have a feeling you'll find quite a few.

[Brianna]

You could just look at every 8800GT Crysis benchmark and review that's online. I have a feeling you'll find quite a few.

There could be jump in speed in windows using the Mac because of FB-DIMMS. Crysis was developed with multiple processors in mind and because the Mac Pro uses FB-DIMMs the game will be able to reach in and use all 8 cores on a Mac when otherwise that wouldn't normally happen. Although being that FB-DIMMS is slower than high performance unbuffered memory of the same size, the game could still be slower. It will be neat to see the outcome. Maybe because of Skulltrail we might se someone develop some higher performance FB-DIMM's in the future. I'd like to hope so. :)

[Eidorian]

There could be jump in speed in windows using the Mac because of FB-DIMMS.Every benchmark I've seen using FB-DIMM has shown poorer performance when compared to standard DDR2 RAM.

Crysis was developed with multiple processors in mind and because the Mac Pro uses FB-DIMMs the game will be able to reach in and use all 8 cores on a Mac when otherwise that wouldn't normally happen. Although being that FB-DIMMS is slower than high performance unbuffered memory of the same size, the game could still be slower. It will be neat to see the outcome.I haven't seen any scaling beyond 2.2-2.4 GHz on a Core 2 Duo. After that the game is GPU dependent. There isn't a quad core benefit, yet.

Maybe because of Skulltrail we might se someone develop some higher performance FB-DIMM's in the future. I'd like to hope so. :)Skulltrail is stillborn with no AMD competition in even the single quad core processor market.

[Brianna]

I haven't seen any scaling beyond 2.2-2.4 GHz on a Core 2 Duo. After that the game is GPU dependent. There isn't a quad core benefit, yet.

That's because you cant use all 8 cores without FB-DIMMS. None has FB-DIMMS in a Machine of that caliber other than Apple.
HardOCP tested a skulltrail (which uses FB-DIMMS) with Crysis and it did use all 8 processor cores.

[Freyqq]

That's because you cant use all 8 cores without FB-DIMMS. None has FB-DIMMS in a Machine of that caliber other than Apple.
HardOCP tested a skulltrail (which uses FB-DIMMS) with Crysis and it did use all 8 processor cores.

FB-DIMM just means that it's buffered ram. I don't see what being FB-DIMM has to do with using all 8 cores. It's benefits are for reliabilty for server applications..its not really designed to be fast. I agree with the above guy..it should be a tad slower than DDR2.

[Eidorian]

That's because you cant use all 8 cores without FB-DIMMS. None has FB-DIMMS in a Machine of that caliber other than Apple.
HardOCP tested a skulltrail (which uses FB-DIMMS) with Crysis and it did use all 8 processor cores.http://www.extremetech.com/article2/0,1558,2209093,00.asp

I'd be more concerned about the FSB instead of the RAM on a multicore and multisocket system.

[stoppard]

I played the crysis demo on my mac pro with 8800 GT at 1024X768 and subjectively the gameplay was smooth with recommended settings (i think high on everything). I didn't look into things like framerates and what not. I just played and my experience was the levels loaded fast and the computer didn't get in the way of me virtually murdering lots and lots of enemy soldiers.

[Mrbill317]

I hope it runs well

[[G5]Hydra]

FB-DIMMS are more about making the mother board a little simpler and having to lay out less traceroutes on a professional duty machine like a server. As you add to the RAM capacity with regular RAM you need more and routes set up on the motherboard. With 2-4GB in a consumer machine this is of no concern but when you start going to 16-32-64 GB etc this becomes a major concern. The FB-DIMMS make the motherboard design much simpler but you do pay a price in latency which is why it takes a small hit compared to regular DDR stuff of equal speed. The Mac Pro uses Xeon's though which have bigger caches among other things so performance in something like games will prob still be the same or maybe even slightly better than using regular Core Duo 2 chips.

[sdschwendener]

As home basic only allows use of one cpu, how much of a hit would performance take in crysis running home basic 64 bit as opposed to business 64-bit which allows the os to use the 2nd xeon.

[MrPDaddyHimself]

That's a PC only game correct?

[knome]

As home basic only allows use of one cpu, how much of a hit would performance take in crysis running home basic 64 bit as opposed to business 64-bit which allows the os to use the 2nd xeon.

1 cpu to 4 cpu's makes huge difference. That would make your cpu your bottle neck and probably cut your potential fps by 1/3 to 1/2. The 64-bit also helps a whole lot. I'm running 64-bit business currently.

[aaronw1986]

That's a PC only game correct?

Yes

[exspes]

This is one of my first orders of business when my mac gets here. Of course, that might be a few days, and another day or so to get vista all prepped for it. So be patient, I'll test that, UT3, Gears, Bioshock, all that stuff.

[XianPalin]

Every time I see new posts on this thread I come in here hoping someone actually did some Crysis playing and posted some results :( :p

[SolrFlare]

Results:

I can run Crysis on Vista Ultimate(64bit) with the latest nvidia drivers at 30fps in most(but not all areas) at:

1680x1050 resolution
No AA
Vsync enabled
Every setting except Shadows and Shaders on Very high
Shadows and Shaders on Medium

My Early 2008 Mac Pro is 4gb of Ram quad core(not 8 core).

I can tell you this too, using r_displayinfo = 1 to display the FPS as well as memory/CPU consumption:

The bottleneck is still the graphics card, not anything else in our system. I've got plenty of room CPU wise and the game itself only uses, at most 1.8gb of Ram. This is further evidenced by Shaders/Shadows/AA being the biggest hit while textures, physics, etc are like a blink of the eye to the mac pro.

[davewolfs]

Additional CORES won't do anything on this game now or EVER. A higher clocked single CPU will run this game faster then a slower clocked 8 core setup.

[barefeats]

Is there a timedemo file for Crysis similar to what's available for Quake 4 and Prey and Doom 3? We'd like to add it to our suite of 3D accelerated game benchmarks.

[Cindori]

You will not be able to get even close to maxing the game. I have a "old" Mac Pro with 2gig ram and a overclocked 8800GTX (wich is alot better), and I run on 1600x1200, very high settings, 2 aa with bout 15 fps. No aa and 1680x1050 I get playable rate @ 35. With 8aa i get 5 fps. With 16aa i get 0,6 fps and game crasches. Remember that ppl with Four 8800Ultra in SLI are barely getting a nice framerate on maximum settings.

[mikekim75]

There could be jump in speed in windows using the Mac because of FB-DIMMS. Crysis was developed with multiple processors in mind and because the Mac Pro uses FB-DIMMs the game will be able to reach in and use all 8 cores on a Mac when otherwise that wouldn't normally happen. Although being that FB-DIMMS is slower than high performance unbuffered memory of the same size, the game could still be slower. It will be neat to see the outcome. Maybe because of Skulltrail we might se someone develop some higher performance FB-DIMM's in the future. I'd like to hope so. :)

I would really like to know where it says that FB Dimms are slow, ALL my searches say that FB Dimms are fast. In fact since they run series and not in parallel they are much faster than DDR 2, also FB Dimms are DDR2 with other stuff. I just want to understand a little better, I'm not arguing. Thanks

[zmunkz]

...In fact since they run series and not in parallel they are much faster than DDR 2,...

I don't know what the answer is here, but in general terms wouldn't parallel be faster than series?

[Brianna]

FB-DIMM just means that it's buffered ram. I don't see what being FB-DIMM has to do with using all 8 cores. It's benefits are for reliabilty for server applications..its not really designed to be fast. I agree with the above guy..it should be a tad slower than DDR2.

Read the article, and also research the benefits of FB-DIMMs. FB-DIMMs are slower than high performance Non Buffered DIMMS, but they have multiprocessor advantages also. FB-DIMMS are not intended for gaming. Usually you''ll see about a 40% decrease in speed with FB-DIMMS vs. High performance non buffered DIMMS with the exact same spec. Although with Skulltrail that isn't always the case. I think it was Toms Hardware that did another review on it that went ballistic over the use of FB-DIMMs over non buffered memory.

Crysis boasts multi-core implementation and we are actually seeing it at this low resolution. Obviously we are not GPU limited and given the use of FB-DIMMs, Crysis will in fact reach into 8 processor cores given the opportunity.

http://enthusiast.hardocp.com/article.html?art=MTQ1OCw1LCxoZW50aHVzaWFzdA==

[knome]

Additional CORES won't do anything on this game now or EVER. A higher clocked single CPU will run this game faster then a slower clocked 8 core setup.
I wave the BS flag. This game has support for up to 4 cpu's. I know several people who went from a core 2 duo to the new quad cores and notices a huge increase in fps. 15-18 fps gain.

[knome]

I would really like to know where it says that FB Dimms are slow, ALL my searches say that FB Dimms are fast. In fact since they run series and not in parallel they are much faster than DDR 2, also FB Dimms are DDR2 with other stuff. I just want to understand a little better, I'm not arguing. Thanks

The whole idea that its fully buffered and error correcting. So it has to check over everything before it can be used. Adding lag time.

[Eidorian]

I wave the BS flag. This game has support for up to 4 cpu's. I know several people who went from a core 2 duo to the new quad cores and notices a huge increase in fps. 15-18 fps gain.http://forums.macrumors.com/showpost.php?p=4940544&postcount=7

What about posting a link to benchmarks of the patched version of Crysis on different processors?

Crysis v1.1 Patch Changelog:

Fixed:

* Potential crash in D3D10
* Orange boxes apearing when hispec savegame loaded into lowspec game.
* Inconsistent damage dealt to vehicles when shot by LAW.
* Reflection resolution on D3D10, MultiGPU reflection update fix
* Memory leak with FSAA modes
* Infinite ammo hacks.
* Memory leak in D3D10 when switching screen modes
* [Multi Core?] optimizations
* When player melees during gun raise animation, their gun will be in a permanently raised position.
* Crash when loading savegame with level exported recently by editor
* Virtual keyboard does not function properly when a game pad is connected
* Users can lose the ability to look around with the Right Stick
* Setting screen resolution to “default� stops user from selecting last resolution
* Bug when changing resolution in D3D10
* Issues with Depth of field and water droplets in D3D10
* Crash on NaN warning

Updates:

* Added: Motion Blur UI and V.SYNC UI options
* Optimized: Motion blur
* Optimized: FSAA (Full Scene Anti-Aliasing)
* Optimized sound id implementation
* Enabled VSync functionality in D3D10
* New benchmarking files for ice CPU benchmark.
* http/xmlrpc password protected remote control session
* Marked debug cvars as cheat
* F12 (screenshot) now works in restricted mode as well

Tweaks:

* Reduced LAW splash damage vs. infantry in PowerStruggle mode
* Slowed Rocket projectile speed down in MP slightly
* Disabled automatic turret bounding boxes on vehicles to prevent issues with LAW hit detection
* Reduced grenade explosion radius in multiplayer
* Clamped water tesselation to avoid cheating in MP
* MultiGPU improvements with depth map updates

[mikekim75]

The whole idea that its fully buffered and error correcting. So it has to check over everything before it can be used. Adding lag time.

This I can understand much better than mumbo jumbo tech talk, If it has to do more work before it's used then of corse it will be a little slower. But how noticeable is it I wonder. I'll have to spend another 3 or 4 grand and find out on my own.:eek:

[kirkbross]

So what kind of frame rate (load) can you with all settings maxed?

[Cindori]

You will not be able to max out Crysis on a Mac Pro.


I dont know if it's because Mac users usually dont read up on hardware, but the 8800GT is actually a pretty weak card, considering 9800 is weeks away, and that demanding games like Crysis require at least 2 or 3 SLI'd GTX/Ultra for maximum experience..

[slackpacker]

Results:

I can run Crysis on Vista Ultimate(64bit) with the latest nvidia drivers at 30fps in most(but not all areas) at:

1680x1050 resolution
No AA
Vsync enabled
Every setting except Shadows and Shaders on Very high
Shadows and Shaders on Medium

My Early 2008 Mac Pro is 4gb of Ram quad core(not 8 core).

I can tell you this too, using r_displayinfo = 1 to display the FPS as well as memory/CPU consumption:

The bottleneck is still the graphics card, not anything else in our system. I've got plenty of room CPU wise and the game itself only uses, at most 1.8gb of Ram. This is further evidenced by Shaders/Shadows/AA being the biggest hit while textures, physics, etc are like a blink of the eye to the mac pro.

You Have 4 GB But BootCamp will only see 2GB..... is this true??

[slackpacker]

I have a 8 Core 2.8 with ATI 2600 XT. If I do click the optimal settings button it sets it to medium at 1024x768. Everything runs smooth as silk.

NO AA if you put it on a 2x it slows down quite a bit.

[Tallest Skil]

You Have 4 GB But BootCamp will only see 2GB..... is this true??

32-bit Vista/XP will only see 3.3GB. 64-bit will see it all and more.

[exspes]

You will not be able to max out Crysis on a Mac Pro.


I dont know if it's because Mac users usually dont read up on hardware, but the 8800GT is actually a pretty weak card, considering 9800 is weeks away, and that demanding games like Crysis require at least 2 or 3 SLI'd GTX/Ultra for maximum experience..

From what I read, the 8800GT is a pretty powerful card, only a few frames slower than 8800GTX in real world tests.

Whether it's powerful enough to run Crysis on max is a whole another issue, and no, it's not.

Then again, you'll be able to play Crysis better than on a lot of computers...

[Eidorian]

You Have 4 GB But BootCamp will only see 2GB..... is this true??On the Mac Pro and any Santa Rosa based Macs you'll have 2 GB for applications and 2 GB for the kernel under 32-bit Windows.

[Lurgen]

You will not be able to max out Crysis on a Mac Pro.


I dont know if it's because Mac users usually dont read up on hardware, but the 8800GT is actually a pretty weak card, considering 9800 is weeks away, and that demanding games like Crysis require at least 2 or 3 SLI'd GTX/Ultra for maximum experience..

It still amazes me how many people say this with little or no factual basis for their comments. I recently finished Crysis on my old PC, and while my Mac Pro hasn't turned up I can at least give specific feedback on what hardware can and can't run the game in a useful way.

I had quality settings bumped up to "High" in almost every category, with AA enabled at 1680x1050x32 without problems. In intense action scenes there was occassional slow-down, but only briefly. The hardware? No SLI, just one (not three) nVidia GeForce 8800GTS and a Core2 Duo E6600 CPU.

Crysis does NOT require 3 SLI'd GTX/Ultra cards for anything but 100% maxed out settings under Vista at resolutions that nobody actually plays at. If you're going to play it on a normal monitor, and don't see any difference between 30fps and 100fps (which the human eye cannot see) then SLI isn't going to improve your experience.

The 8800GT has consistently been benchmarked as being within 5% (+ or -, depending on the test-cases) of the 8800GTS. No, it isn't the fastest card on the market, but it also isn't any slouch. Sure, the 9000 series of card will be out any month now, but they aren't going to make an immense difference to the average gamer. Not that the average gamer buys a Mac Pro, but even if they did chances are they're running Vista or XP on it so adding a better video card is no real problem.

It is also important to remember that a lot of the perceptions regarding Crysis at max settings relate to Vista. You can't max out the settings in Crysis without DirectX10, which in turn requires Vista (OK, yes there are hacks to make it look like you have max settings under XP but you're actually running DX9 emulations of the DX10 features, which is not the same thing). Vista sucks a large percentage of your performance when compared to XP, and significantly increases your hardware requirements. World of Warcraft, for example, ran at a constant 100fps under XP but regularly dipped below 30fps under Vista on the same hardware. Needless to say, Vista didn't get to stay.

[elvisizer]

and anyway, the mac pro's support ANY pc video card under windows.
If you really need max FPS, you can always upgrade with a standard PC video card and keep the 8800gt for OS X only.
I'll probably do that eventually, but the 8800gt is fine for now. I'm coming from an AMD fx-60 with dual 7800gtx's in SLI. A single 8800gt beats that set up.

[slackpacker]

Dudes you all need to see this ....especially LURGEN, Don't knock it until you try it.

http://forums.macrumors.com/showthread.php?t=427751

Even on a 2600XT the game looks Beautiful :D!

[Mercuric Oxide]

Dudes you all need to see this ....especially LURGEN, Don't knock it until you try it.

http://forums.macrumors.com/showthread.php?t=427751

Even on a 2600XT the game looks Beautiful :D!

I think you mean Cindori

[Inlakech]

I tried Crysis on my 2,8 Octo with 16 GB Ram on WinXp64 yesterday. But I didnt get it running in 64bit mode, there was a secuROM error ... Does anybody know about this issue ?

I took these settings:

1920 x 1080, 2 AA , everything set to high (when I hit the optimize button it goes to high aswell) ... The image quality on my 30" Cinema Display blew me away :eek: :eek: :eek: . Looks like a movie. Plays very smooth :D . I didnt look up the framerates yet.

[slackpacker]

I think you mean Cindori

Nope Lurgen sort of said it was not worth doing your Crysis fix. I totally changed the game for me.

Lurgen Said "It is also important to remember that a lot of the perceptions regarding Crysis at max settings relate to Vista. You can't max out the settings in Crysis without DirectX10, which in turn requires Vista (OK, yes there are hacks to make it look like you have max settings under XP but you're actually running DX9 emulations of the DX10 features, which is not the same thing). "

[Cindori]

Well Lurgen, I guess you were playing at XP and DX9.

[aaronw1986]

I tried Crysis on my 2,8 Octo with 16 GB Ram on WinXp64 yesterday. But I didnt get it running in 64bit mode, there was a secuROM error ... Does anybody know about this issue ?

I took these settings:

1920 x 1080, 2 AA , everything set to high (when I hit the optimize button it goes to high aswell) ... The image quality on my 30" Cinema Display blew me away :eek: :eek: :eek: . Looks like a movie. Plays very smooth :D . I didnt look up the framerates yet.

Do you own a copy? This sounds like a pirating issue, the secuROM

[bpd115]

and don't see any difference between 30fps and 100fps (which the human eye cannot see) then SLI isn't going to improve your experience.



This is one of the biggest myths out there.

You most definitely can tell the difference between 30 fps and 60 fps.

While 30 fps is by all means fine, 60 fps is smoother.

Look at this years NCAA Football which ran at 60 FPS on the 360 and 30 on the PS3.

Watch the following:

http://loot-ninja.com/2007/04/29/video-comparison-24fps-vs-60fps/

and to quote from:

http://www.daniele.ch/school/30vs60/30vs60_1.html

"So what is the answer to how many frames per second should we be looking for? Anything over 60 fps is adequate, 72 fps is maximal (anything over that would be overkill). Framerates cannot drop though from that 72 fps, or we will start to see a degradation in the smoothness of the game. Don't get me wrong, it is not bad to play a game at 30 fps, it is fine, but to get the illusion of reality, you really need a frame rate of 72 fps. "

[Sesshi]

That's because you cant use all 8 cores without FB-DIMMS. None has FB-DIMMS in a Machine of that caliber other than Apple.

No, Dell and HP do for starters.

I have an 8800 related question - would the 8800GTX / 8800GTX Ultra have any possibility of working on the Pro now?

[knome]

Yea 30 fps is the limit of what is blatantly obvious. That is where we can identify each individual frame. The U.S. Military did a test and i believe they said people could notice the difference up to 250fps. All they noticed was a smoothness to the picture that was not present before.

[killmoms]

Yea 30 fps is the limit of what is blatantly obvious. That is where we can identify each individual frame. The U.S. Military did a test and i believe they said people could notice the difference up to 250fps. All they noticed was a smoothness to the picture that was not present before.

Exactly, it's not that you can perceive every individual frame's change at 250fps, but it definitely looks smoother than, say, 100fps. This is especially true for real-time computer generated imagery, since it lacks the motion blur inherent to film or television. Pause a movie and you'll see that a high-motion frame is very blurred because of the film being exposed for a fraction of a second in order to receive enough light to develop properly. Because of this effect, movies at 24 frames per second are very watchable because this captured motion blur helps "blend" the frames temporally in our brains.

In comparison, a frame of a video game being rendered by a graphics card is "exposed" instantaneously (because calculating motion blur is very expensive in terms of CPU/GPU time) and lacks this blur. Because of that you will notice a drop or increase in frame rate much more.

[student_trap]

Watch the following:

http://loot-ninja.com/2007/04/29/video-comparison-24fps-vs-60fps/


hmmm, i just watched this, and maybe its because of the size of the video (its tiny!0, but I really couldn't tell the difference!

It could I suppose also just be my eyes!

[Sesshi]

Also, an average framerate is that - average. Some scenes could have higher or lower framerates? On my SLI'd 8800 systems I rarely have issues but there are occasional moments in demanding games, especially when I jack everything up, that it does go down in certain situations.

I find what that happens, it's very distracting even if the game is still perfectly playable.

[Inlakech]

Do you own a copy? This sounds like a pirating issue, the secuROM

I bought it. I own the original DVD, no copy.

I guess it has to do or caused a windows xp64 Installer Service Error. I cannot install any software on windows anymore. Need to reinstall completely.

[Firefly2002]

That's because you cant use all 8 cores without FB-DIMMS. None has FB-DIMMS in a Machine of that caliber other than Apple.
HardOCP tested a skulltrail (which uses FB-DIMMS) with Crysis and it did use all 8 processor cores.

http://www.tomshardware.com/2008/02/08/intel_skulltrail_part_3/page6.html

Sure about that?

Apple's machines aren't particularly powerful in comparison to what you could build yourself were you so inclined. I could build a Quad-core PC system that ran circles around the 8-core 3.2 GHz Mac Pro in almost every instance, save where it really took advantage of all eight cores. And if I wanted, I could have it do it on Mac OS X. Intel Cores are nice overclockers... speeds reaching 4 GHz are far from unheard of, and the extra memory bandwidth I'd get out of much nicer DDR2-800 DIMMs (or better) would help more. When it came to gaming, of course, the Mac Pro wouldn't be able to compete at all. I could drop in about 8 GB DDR, 2x1TB drives, and have a killer system for less than the Mac Pros.

[[G5]Hydra]

There seems to be some confusion about FB-DIMMs and why they are being used in the Mac Pro. There is no performance advantage in using FB-DIMMs, and in fact you get slightly less performance compared to regular DDR. FB-DIMMs are used by intel on their high end workstation/server boards for one reason only, they make it far easier and cheaper to allow mass quantities of RAM. In computational situations where you need 16, 32, 64GB or more of memory absolute speed may be secondary to having enough RAM available.

Performance is slightly worse than regular DDR memory because of the added latency each module will add when in place. Each FB-DIMM has what is called an Advanced Memory Buffer (AMB) which is essentially a small memory controller. Most other computers have a memory controller on the motherboard or the main CPU which essentially controls the memory directly. This has an added advantage of less latency because of the lack of go betweens. The disadvantage is as you add more sockets for more memory you need more pins on the memory controller and more wires on the motherboard. Adding more and more RAM capacity makes the problem get worse.

FB-DIMMS don't talk to the memory controller directly they have the AMB on each stick which gets the request first and also acts as a repeater for the requests that go downstream. The added benefit is that there are some fault corrections built in that impose no performance hit compared to a regular DDR based system. Theoretically a system based on the FB-DIMM architecture would have higher bandwidth and be cheaper and easier to manufacture with a much less complicated motherboard. The downside is all the AMB's inline will increase latency and they get hot so they will also introduce more heat into a computer that uses them. The price difference is mostly due the the low volume they are made compared to regular DDR, in fact the memory chips on a FB-DIMM theoretically can be anything as the AMB would be the controller that interfaces with the rest of the system and are really just regular memory chips. The good news on price is that intel's new Skull Trail if popular should make the FB-DIMM more popular and help bring down the prices.

As it stands today the reasons intel chose to use them in the server/workstation arena are to allow very large memory capacity, cheaper and easier to manufacture motherboards, and some better fault tolerance and correction which would appeal to professional, scientific and other high end usages.

In everyday use a system based on FB-DIMMs will likely be a little slower than a standard DDR one but it should be pretty close in any event. Mac Pros use Xeons which are essentially souped up Core Duo 2's with a few nice features which should actually more than make up for the latency associated with the FB-DIMMs, especially the brand new ones. The new Xeons have 12 MB L2 cache per processor (6 MB shared per pair of cores) and a 24 MB snoop filter on the FSB which essentially keeps track of all the cached data on both CPU's which should dramatically reduce traffic on the FSB and push bandwidth usage to the limits. There are quite a nice little bunch of tweaks in there.

[almostinsane]

http://www.tomshardware.com/2008/02/08/intel_skulltrail_part_3/page6.html

Sure about that?



What are you quoting? That Skulltrail still uses FB-DIMMS?

[slackpacker]

32-bit Vista/XP will only see 3.3GB. 64-bit will see it all and more.


Thanks... this is great info! So much misinformation going around. Does 64 bit help at all with Crysis?

[Mercuric Oxide]

Nope Lurgen sort of said it was not worth doing your Crysis fix. I totally changed the game for me.

Lurgen Said "It is also important to remember that a lot of the perceptions regarding Crysis at max settings relate to Vista. You can't max out the settings in Crysis without DirectX10, which in turn requires Vista (OK, yes there are hacks to make it look like you have max settings under XP but you're actually running DX9 emulations of the DX10 features, which is not the same thing). "

Oh, I see. Thanks for defending me :D

[Freyqq]

This is one of the biggest myths out there.

You most definitely can tell the difference between 30 fps and 60 fps.

While 30 fps is by all means fine, 60 fps is smoother.

Look at this years NCAA Football which ran at 60 FPS on the 360 and 30 on the PS3.

Watch the following:

http://loot-ninja.com/2007/04/29/video-comparison-24fps-vs-60fps/

and to quote from:

http://www.daniele.ch/school/30vs60/30vs60_1.html

"So what is the answer to how many frames per second should we be looking for? Anything over 60 fps is adequate, 72 fps is maximal (anything over that would be overkill). Framerates cannot drop though from that 72 fps, or we will start to see a degradation in the smoothness of the game. Don't get me wrong, it is not bad to play a game at 30 fps, it is fine, but to get the illusion of reality, you really need a frame rate of 72 fps. "

a typical display runs at 60 hertz. Therefore, it flashes frames at 60 fps. If the game runs above 60 fps, the monitor will only display at 60 fps anyways....i don't get it.

[Mercuric Oxide]

Are you getting the 8800 slackpacker?

Anyway, AMD released new drivers today, I haven't testes them yet, but people are reporting a 10 fps increase in Crysis.

Let me know if you try it

http://game.amd.com/us-en/drivers_catalyst.aspx?p=xp/radeonx-xp

(for xp pro)

[slackpacker]

Are you getting the 8800 slackpacker?

Anyway, AMD released new drivers today, I haven't testes them yet, but people are reporting a 10 fps increase in Crysis.

Let me know if you try it

http://game.amd.com/us-en/drivers_catalyst.aspx?p=xp/radeonx-xp

(for xp pro)

Cool Thanks!

Yes I have the 8800 on order--- March 31st is my delivery date HA

Best Regards

Update~ I got the Mac Pro on the 12th and Got Crysis the day of the Driver release on the 13th :). So thats the first driver I installed. I have been using that all along. Works great as far as I can see. No Crashes and everything runs really well. ATI Driver ver 8.2

[Mercuric Oxide]

Cool Thanks!

Yes I have the 8800 on order--- March 31st is my delivery date HA

Best Regards

Update~ I got the Mac Pro on the 12th and Got Crysis the day of the Driver release on the 13th :). So thats the first driver I installed. I have been using that all along. Works great as far as I can see. No Crashes and everything runs really well. ATI Driver ver 8.2

coolness

[exspes]

Alright guys, as promised -

I only have the Crysis demo, so the final game might be more optimized, especially with 1.1. I'm also running 32bit Vista Ultimate - which shouldn't matter since I only have the stock 2gb of RAM at the moment. But I do have access to all 8 cores.

So, maxed out Crysis: 1920x1200, everything on Very High. No AA. Getting 10-12 frames per second all around. That's "choppy but playable for single player" in my book. Firefights are obviously worse.

Going to play around with some settings and post more results.

EDIT: some more:

Knocking down resolution from 1920x1200 to 1680x1050 didn't do... anything at all to the framerate.
I get about 8-10 fps in open areas overlooking water.

[slackpacker]

Alright guys, as promised -

I only have the Crysis demo, so the final game might be more optimized, especially with 1.1. I'm also running 32bit Vista Ultimate - which shouldn't matter since I only have the stock 2gb of RAM at the moment. But I do have access to all 8 cores.

So, maxed out Crysis: 1920x1200, everything on Very High. No AA. Getting 10-12 frames per second all around. That's "choppy but playable for single player" in my book. Firefights are obviously worse.

Going to play around with some settings and post more results.

EDIT: some more:

Knocking down resolution from 1920x1200 to 1680x1050 didn't do... anything at all to the framerate.
I get about 8-10 fps in open areas overlooking water.

I guess this is with the 8800GT Card?

[exspes]

I guess this is with the 8800GT Card?

Hah, yep.

Interesting thing is, I went through "high" "medium" and "low settings" after having run it on very high.

Very high - 8-12fps
High - 9-13fps
Medium - 14-20fps
Low - 28-32fps

This was all at 1920x1200

[Brianna]

Hah, yep.

Interesting thing is, I went through "high" "medium" and "low settings" after having run it on very high.

Very high - 8-12fps
High - 9-13fps
Medium - 14-20fps
Low - 28-32fps

This was all at 1920x1200

What processor(s), single or dual, are you running, and how much memory do you have, and what speed is that memory running?

[slackpacker]

What processor(s), single or dual, are you running, and how much memory do you have, and what speed is that memory running?

He said 2gb and I assume its a 8 core (2 processor) 2.8ghz

[exspes]

What processor(s), single or dual, are you running, and how much memory do you have, and what speed is that memory running?

Sorry, I should have probably provided full system specs

2.8Ghz Octo-Core
2gb of 800mhz ram
running vista 32bit Ultimate
8800GT card with latest drivers
Running Crysis demo


Different game: I ran the UT3 beta/demo - didn't get a formal fps count, but it felt smooth on any map I've played. And by smooth I am guessing well over 30fps 1920x1200 every setting cranked. That means chances are you can play any of the UE3 blockbusters really well (Bioshock, Mass Effect, etc)

[Mercuric Oxide]

Alright guys, as promised -

I only have the Crysis demo, so the final game might be more optimized, especially with 1.1. I'm also running 32bit Vista Ultimate - which shouldn't matter since I only have the stock 2gb of RAM at the moment. But I do have access to all 8 cores.

So, maxed out Crysis: 1920x1200, everything on Very High. No AA. Getting 10-12 frames per second all around. That's "choppy but playable for single player" in my book. Firefights are obviously worse.

Going to play around with some settings and post more results.

EDIT: some more:

Knocking down resolution from 1920x1200 to 1680x1050 didn't do... anything at all to the framerate.
I get about 8-10 fps in open areas overlooking water.


Hmm, that's disappointing, but I'll assume that the final game along with 1.1/2 and latest nvidia drivers will provide better results.

[contoursvt]

Hmmm on my 8800GT I was averaging around 50fps or so on very high but at a low resolution of 1024x768 on a 21" CRT (its on a quad core PC running vista64)

If I remember right, bumping it up to 1280x1024 lowered the framerate by about 10fps. I can try tonight and see what I get at different resolutions but I have no widescreen displays so I'll have to try 4:3

[exspes]

Hmmm on my 8800GT I was averaging around 50fps or so on very high but at a low resolution of 1024x768 on a 21" CRT (its on a quad core PC running vista64)

If I remember right, bumping it up to 1280x1024 lowered the framerate by about 10fps. I can try tonight and see what I get at different resolutions but I have no widescreen displays so I'll have to try 4:3

I'll try a significantly lower resolution later on. I like running games at native resolution (I'd rather knock detail down) for some reason, but I'll see what's up on lower ones.

[contoursvt]

Ya if I had an LCD, I"d feel the same as you do, but since I'm still using my old school CRT monitors (they still work great), there is no 'native' resolution so no scaling issues. 1024 looks great on very high by the way :)

I'll try a significantly lower resolution later on. I like running games at native resolution (I'd rather knock detail down) for some reason, but I'll see what's up on lower ones.

[thagomizer]

Hah, yep.

Interesting thing is, I went through "high" "medium" and "low settings" after having run it on very high.

Very high - 8-12fps
High - 9-13fps
Medium - 14-20fps
Low - 28-32fps

This was all at 1920x1200

Run it with SHADERS and SHADOWS set to medium, everything else set to high. It's playable this way on my MacBook Pro w/ 8600M GT (256MB VRAM), so I'm sure you'll have no problem with an 8800GT.

[elvisizer]

So, maxed out Crysis: 1920x1200, everything on Very High. No AA. Getting 10-12 frames per second all around. That's "choppy but playable for single player" in my book. Firefights are obviously worse.


yeesh, i guess different people have different thresholds for perceiving frame drops. I installed crysis lasst night on my pro (10gb RAM, 8800gt, vista 64bit). everything on very high was totally unplayable. I settled on everything on high, except for shaders, shadows, and physics set to medium, no AA, 1920x1200. I didn't turn on the FPS display, but with those settings, the motion blur when you whip the mouse around quickly is super smooth.
I couldn't see much of a quality difference from reducing shaders and shadows. Actually, i thought it looked a bit better!

[ray_uk]

:D

Is anyone here running it with either a 8x1GB setup or a 4x2GB of RAM setup, from what i've read so far, if the Dimms are distributed evenly across both risers then there is a performance advantage due to the Mac Pro's RAM running in a similar mode to dual channel on a pc? Not been able to confirm this but would be good to know.

:D

[elvisizer]

:D

Is anyone here running it with either a 8x1GB setup or a 4x2GB of RAM setup, from what i've read so far, if the Dimms are distributed evenly across both risers then there is a performance advantage due to the Mac Pro's RAM running in a similar mode to dual channel on a pc? Not been able to confirm this but would be good to know.

:D

the mac pro has 4 memory channels. optimal config is 4 dimms of whatever capacity.

[Cindori]

Well

A single 8800GT is far from enough to crank up Crysis to very high in DX10. If you are getting high FPS you are probably running XP, and then Crysis is not so much different from any other game out there.

[elvisizer]

remember, everyone, don't feed the trolls. :p

[Brianna]

Sorry, I should have probably provided full system specs

2.8Ghz Octo-Core
2gb of 800mhz ram
running vista 32bit Ultimate
8800GT card with latest drivers
Running Crysis demo


Different game: I ran the UT3 beta/demo - didn't get a formal fps count, but it felt smooth on any map I've played. And by smooth I am guessing well over 30fps 1920x1200 every setting cranked. That means chances are you can play any of the UE3 blockbusters really well (Bioshock, Mass Effect, etc)

Are you sure your using the latest drivers? Did you install "Leopard Graphics Update 1.0" ? This is a separate update after the 10.5.2 update. You might want to run software update one more time if you didn't install Leopard Graphics Update 1.0 yet.

[exspes]

Are you sure your using the latest drivers? Did you install "Leopard Graphics Update 1.0" ? This is a separate update after the 10.5.2 update. You might want to run software update one more time if you didn't install Leopard Graphics Update 1.0 yet.

Was the leopard graphics update firmware? Otherwise it wouldn't affect windows. However, I'll be installing a 64bit windows vista tomorrow, and then i'll manually go through and ensure my drivers work. After all that I'll run the game again and do it at many different resolutions with many settings. This will take me some time though. I'll try to get a nice writeup going sometime next week.

[Brianna]

Don't knock it until you try it. Maybe there is a firmware patch in the update. Who would know but Apple.
As far as I can tell you're the only one having this problem. I would consult help in a PC forum. I'm absolutely sure PC people have more experience diagnosing PC problems.
Other than that how is your hardware running on the Mac side?

[slackpacker]

Don't knock it until you try it. Maybe there is a firmware patch in the update. Who would know but Apple.
As far as I can tell you're the only one having this problem. I would consult help in a PC forum. I'm absolutely sure PC people have more experience diagnosing PC problems.
Other than that how is your hardware running on the Mac side?

Leopard Graphics Driver Update (hence the name) - Is ONLY updating drivers for OS X. No firmware update as well. But there is an Apple updater app in your Bootcamp partition for Apple Updates.

Once in Bootcamp you need to goto nVidia and d/l the latest drivers. Apple is usually quite behind on Windows Graphics Drivers. So if you did not install those you will probably have issues.

Also the Demo is does not contain the 1.1 update for Crysis which increases performance as well.

I don't think your having a problem, your just setting the bar to high for a game that is much more detailed than the other games you are comparing it to.

Try running it a 1280x800 resolution. Also turn down some of the graphic options to medium. Even a workhorse like the Mac Pro with a single non SLI Graphics card is no match for the Ultrahigh settings in Crysis

[Brianna]

Leopard Graphics Driver Update (hence the name) - Is ONLY updating drivers for OS X. ..............

The name is not Leopard Graphics "Driver" update. Hence the possibility of anything graphically involved.

Originally posted @Apple.com
Leopard Graphics Update 1.0

About Leopard Graphics Update
Leopard Graphics Update is recommended for all users and improves the stability and compatibility of your Mac. This update requires Mac OS X 10.5.2.

[slackpacker]

The name is not Leopard Graphics "Driver" update. Hence the possibility of anything graphically involved.

Possible...

But it does not change anything for BootCamp since it only effects the Leopard OS. It can't because OS X normally cannot see or write to an NTFS partition which is BootCamp. If it added a new firmware to the card, so be it. But being up to date on the Graphic Drivers for the windows side is a must.

[Brianna]

Well I do know that when you install Windows on a Mac you have to re-insert your Leopard disk for boot camp stuff. Not that I'm saying there is a boot camp driver in the update, but it does mean that Software update could look for boot camp partitions in your system in the future, and issue a boot camp driver update from the Mac side.

You also have a choice of NTFS partition, or FAT. NTFS can not share files back and forth with windows and OS X but a FAT file partition will allow it. That's why I chose it on my partition. It looses a tad bit of performance, but I need to share files. Otherwise I wouldn't have bothered installing windows.

[slackpacker]

Well I do know that when you install Windows on a Mac you have to re-insert your Leopard disk for boot camp stuff. Not that I'm saying there is a boot camp driver in the update, but it does mean that Software update could look for boot camp partitions in your system in the future, and issue a boot camp driver update from the Mac side.

You also have a choice of NTFS partition, or FAT. NTFS can not share files back and forth with windows and OS X but a FAT file partition will allow it. That's why I chose it on my partition. It looses a tad bit of performance, but I need to share files. Otherwise I wouldn't have bothered installing windows.

Figgered I get this post back on track.

As soon as I get my 8800GT in a month or so :rolleyes:

I'll post some really detailed Tests from the Crysis Speedtest files. Until then all I know is that my ATI 2600XT Rocks for the game. It more than playable at tested with FRAPS min 19, Max 25, Avg 21 and this is running at 1280x960 on a 8 Core MP 2.8Ghz

[ComeOnDieYoung]

deleted

[Brianna]

Figgered I get this post back on track.

As soon as I get my 8800GT in a month or so :rolleyes:

I'll post some really detailed Tests from the Crysis Speedtest files. Until then all I know is that my ATI 2600XT Rocks for the game. It more than playable at tested with FRAPS min 19, Max 25, Avg 21 and this is running at 1280x960 on a 8 Core MP 2.8Ghz

Me too. I'm just waiting for windows to arrive. Had to pay Microslosh for a copy. I should have it by Thursday. Never thought I'd have to buy MS$ software in my life, but with Crysis on the PC, and Mass Effect going to PC I'm definately playing a few games. I also want to use Softimage XSI, and Zbrush 3.1 so I have no choice.

[slackpacker]

I found this Benchmarking tool

It works great and will allow us to have some good benchmarks that are accurate

Crysis Benchmark Tool 1.05 Final

http://downloads.guru3d.com/download.php?det=1791

I ran the test Assault Harbor 4 times here are the results for a 2.8Ghz,8 Core,2GB,2600XT

Please note the Graphical settings area.

2/17/2008 2:52:47 PM - XP
Beginning Run #1 on Map-harbor, Demo-Assault_Harbor
DX9 1024x768, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=4, Time Of Day= 5
Global Game Quality: Custom
Custom Quality Values:
VolumetricEffects=Low
Texture=Medium
ObjectDetail=High
Sound=High
Shadows=Low
Water=Low
Physics=High
Particles=Medium
Shading=Low
PostProcessing=Low
GameEffects=Medium
==============================================================
TimeDemo Play Started , (Total Frames: 4100, Recorded Time: 132.23s)
!TimeDemo Run 0 Finished.
Play Time: 180.15s, Average FPS: 22.76
Min FPS: 15.29 at frame 2492, Max FPS: 36.76 at frame 3201
Average Tri/Sec: 2246078, Tri/Frame: 98688
Recorded/Played Tris ratio: 0.14
!TimeDemo Run 1 Finished.
Play Time: 172.99s, Average FPS: 23.70
Min FPS: 15.29 at frame 2492, Max FPS: 36.76 at frame 3201
Average Tri/Sec: 1789062, Tri/Frame: 75485
Recorded/Played Tris ratio: 0.18
!TimeDemo Run 2 Finished.
Play Time: 169.83s, Average FPS: 24.14
Min FPS: 15.29 at frame 2492, Max FPS: 36.98 at frame 3647
Average Tri/Sec: 1692339, Tri/Frame: 70099
Recorded/Played Tris ratio: 0.19
!TimeDemo Run 3 Finished.
Play Time: 167.42s, Average FPS: 24.49
Min FPS: 15.29 at frame 2492, Max FPS: 36.98 at frame 3647
Average Tri/Sec: 1817518, Tri/Frame: 74216
Recorded/Played Tris ratio: 0.18
TimeDemo Play Ended, (4 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/17/2008 2:52:47 PM - XP

Run #1- DX9 1024x768 AA=No AA, 32 bit test, Quality: Custom ~~ Overall Average FPS: 24.11

[contoursvt]

I ran the 32bit version in Vista64. The 64bit version was slower by about 10% for whatever reason. My DX9 numbers for 1600, 1680 and 1900 resolutions are weird. All around the same. I re-ran those resolutions later again and got the same thing so not sure why they are so close yet in DX10 it changes a lot.

Test was run on default island map on the full version of the game. Time was left unchanged at 9 (I assume that was 9am since there was sunlight) and I set it for 2 passes per setting. Vsync off and it was full screen.

Machine is a Q6600 box at stock 2.4 speed with an 8800GT 512MB card. There is 8GB ram in the box but during the demo It showed just over 1GB used for the game.

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/17/2008 7:25:11 PM - Vista 64


Run #1- DX9 1024x768 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 44.61

Run #2- DX9 1280x720 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 38.73

Run #3- DX9 1280x1024 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 39.17

Run #4- DX9 1400x960 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 37.38

Run #5- DX9 1600x1200 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 29.72

Run #6- DX9 1680x1050 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 29.87

Run #7- DX9 1900x1200 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 29.84


<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/17/2008 7:57:07 PM - Vista 64


Run #1- DX10 1024x768 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 41.86

Run #2- DX10 1280x720 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 36.87

Run #3- DX10 1280x1024 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 37.02

Run #4- DX10 1400x960 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 34.70

Run #5- DX10 1600x1200 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 23.91

Run #6- DX10 1680x1050 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 27.71

Run #7- DX10 1900x1200 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 21.91

[slackpacker]

I ran the 32bit version in Vista64. The 64bit version was slower by about 10% for whatever reason. My DX9 numbers for 1600, 1680 and 1900 resolutions are weird. All around the same. I re-ran those resolutions later again and got the same thing so not sure why they are so close yet in DX10 it changes a lot.

Test was run on default island map on the full version of the game. Time was left unchanged at 9 (I assume that was 9am since there was sunlight) and I set it for 2 passes per setting. Vsync off and it was full screen.

Machine is a Q6600 box at stock 2.4 speed with an 8800GT 512MB card. There is 8GB ram in the box but during the demo It showed just over 1GB used for the game.

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/17/2008 7:25:11 PM - Vista 64


Nice Framerates!

I hope I get those when I get my card! I'm starting to get to areas in the game that really mess with the 2600XT's framerates.

[contoursvt]

I'm sure you will have no probs once in the 8800 arrives :) You will really enjoy the much more steady framerates and the eye candy with everything on high.

[Brianna]

I'm sure you will have no probs once in the 8800 arrives :) You will really enjoy the much more steady framerates and the eye candy with everything on high.

Your not using a Mac Pro correct?

[contoursvt]

Thats right. Not a Mac Pro, but pretty much the same 8800GT so scores should be very similar. Actually I read somewhere that Crysis does not make very good use of anything more than 2 cores so a 4 or 8 core will not be fully used.

Your not using a Mac Pro correct?

[slackpacker]

Your not using a Mac Pro correct?

Copied from contoursvt's post~

Machine is a Q6600 box at stock 2.4 speed with an 8800GT 512MB card. There is 8GB ram in the box but during the demo It showed just over 1GB used for the game.

[diamond.g]

I ran the 32bit version in Vista64. The 64bit version was slower by about 10% for whatever reason. My DX9 numbers for 1600, 1680 and 1900 resolutions are weird. All around the same. I re-ran those resolutions later again and got the same thing so not sure why they are so close yet in DX10 it changes a lot.

Test was run on default island map on the full version of the game. Time was left unchanged at 9 (I assume that was 9am since there was sunlight) and I set it for 2 passes per setting. Vsync off and it was full screen.

Machine is a Q6600 box at stock 2.4 speed with an 8800GT 512MB card. There is 8GB ram in the box but during the demo It showed just over 1GB used for the game. <snip>
Could you retest with AA. I think you scores should stay about the same. IMO, you are just wasting the bandwidth of the card without it.

[OnePumpChump]

Never thought I'd have to buy MS$ software in my life, but with Crysis on the PC, and Mass Effect going to PC I'm definately playing a few games.

So what version of windows did you order?

[Brianna]

So what version of windows did you order?

XP Pro SP2, which is upgradable to any version of vista when I choose to do so.

I didn't go with Vista because of Crysis.

[contoursvt]

Could you retest with AA. I think you scores should stay about the same. IMO, you are just wasting the bandwidth of the card without it.

I tested at 2x and 8x to see the effects. I only did it in DX9 but its to show the percentage decrease in performance.

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 11:47:14 AM - Vista 64

Run #1- DX9 1024x768 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 42.59
Run #2- DX9 1280x720 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 37.16
Run #3- DX9 1280x1024 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 33.28
Run #4- DX9 1400x960 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 31.99
Run #5- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #6- DX9 1680x1050 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #7- DX9 1900x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 12:13:27 PM - Vista 64

Run #1- DX9 1024x768 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 39.43
Run #2- DX9 1280x720 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 33.50
Run #3- DX9 1280x1024 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 28.44
Run #4- DX9 1400x960 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 20.92
Run #5- DX9 1600x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.60
Run #6- DX9 1680x1050 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.78
Run #7- DX9 1900x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.62

[diamond.g]

Nice results! Not as low as I thought. If the card were overclocked I think you could regain the few fps lost. But it does show (for at least 2x AA) that the performance hit really isn't that big. With 8xAA that just isn't enough bandwidth available.

I tested at 2x and 8x to see the effects. I only did it in DX9 but its to show the percentage decrease in performance.

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 11:47:14 AM - Vista 64

Run #1- DX9 1024x768 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 42.59
Run #2- DX9 1280x720 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 37.16
Run #3- DX9 1280x1024 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 33.28
Run #4- DX9 1400x960 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 31.99
Run #5- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #6- DX9 1680x1050 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #7- DX9 1900x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 12:13:27 PM - Vista 64

Run #1- DX9 1024x768 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 39.43
Run #2- DX9 1280x720 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 33.50
Run #3- DX9 1280x1024 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 28.44
Run #4- DX9 1400x960 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 20.92
Run #5- DX9 1600x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.60
Run #6- DX9 1680x1050 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.78
Run #7- DX9 1900x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.62

[bensim]

Not to get off the FPS topic altogether, but I've been playing the Crysis demo on my Mac Pro (3.0, 8800GT), and it gives me a headache. I've tried all kinds of different settings. Even at 1024x768 the blurring when you turn look from side to side gets to me. Maybe I'm just not made to feel comfortable in Windows.

[contoursvt]

Maybe you're sitting too close to the screen. Most people who have issues with headaches or feeling sick are too close. You need to see the frame of your monitor fairly easily from what I've read.

Also I think lowering the setting to medium instead of high will disable the motion blur.

Not to get off the FPS topic altogether, but I've been playing the Crysis demo on my Mac Pro (3.0, 8800GT), and it gives me a headache. I've tried all kinds of different settings. Even at 1024x768 the blurring when you turn look from side to side gets to me. Maybe I'm just not made to feel comfortable in Windows.

[Brianna]

I tested at 2x and 8x to see the effects. I only did it in DX9 but its to show the percentage decrease in performance.

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 11:47:14 AM - Vista 64

Run #1- DX9 1024x768 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 42.59
Run #2- DX9 1280x720 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 37.16
Run #3- DX9 1280x1024 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 33.28
Run #4- DX9 1400x960 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 31.99
Run #5- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #6- DX9 1680x1050 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57
Run #7- DX9 1900x1200 AA=2x, 32 bit test, Quality: High ~~ Last Average FPS: 24.57

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 12:13:27 PM - Vista 64

Run #1- DX9 1024x768 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 39.43
Run #2- DX9 1280x720 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 33.50
Run #3- DX9 1280x1024 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 28.44
Run #4- DX9 1400x960 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 20.92
Run #5- DX9 1600x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.60
Run #6- DX9 1680x1050 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.78
Run #7- DX9 1900x1200 AA=8x, 32 bit test, Quality: High ~~ Last Average FPS: 13.62

You were able to choose DX9 instead of DX10 in vista... That is kewl, but I still think that I will get better performance in XP Pro. We'll see.

[contoursvt]

You might get a few fps at low resolution, but when you start getting to 1680x1050 or more, I think it will be all limited by the power of the card and not the OS at all.

You were able to choose DX9 instead of DX10 in vista... That is kewl, but I still think that I will get better performance in XP Pro. We'll see.

[slackpacker]

Not to get off the FPS topic altogether, but I've been playing the Crysis demo on my Mac Pro (3.0, 8800GT), and it gives me a headache. I've tried all kinds of different settings. Even at 1024x768 the blurring when you turn look from side to side gets to me. Maybe I'm just not made to feel comfortable in Windows.

In the full retail version you can turn off the Bobbing back and forth of the main character. I friend of mine had the same problem with FPS's

Also the 1.1 Patch introduces a way to totally turn off Motion Blur.. or scale it to you needs.

[orpheus1120]

I'm not too much a gamer myself, but I would like to ask, hypothetically what is needed to run Crysis at the best settings possible in a Mac Pro?

[slackpacker]

I'm not too much a gamer myself, but I would like to ask, hypothetically what is needed to run Crysis at the best settings possible in a Mac Pro?

Well the best Mac Pro can offer right now is the 8800GT at 3.2Ghz for the Mac Pro. That is the Level you need to really run it well. Right now Crysis can overload even the Best Computers.

I have a MP 2.8Ghz,2600XT and I'm Playing Crysis at Medium level and Having fun. But If I want to up the Graphics to the High.. it will slow down to 8fps.

[diamond.g]

I'm not too much a gamer myself, but I would like to ask, hypothetically what is needed to run Crysis at the best settings possible in a Mac Pro?
Define best settings possible. If it is what I think you are referring to you would need something similar to quad SLI because Crysis puts a big hurt on GPU's.

[orpheus1120]

Define best settings possible. If it is what I think you are referring to you would need something similar to quad SLI because Crysis puts a big hurt on GPU's.

Yes I am referring to what you are thinking. Then perhaps running the best Crysis settings (everything on high) on a Mac Pro is impossible given the Mac Pro doesn't have sli capability (IIRC). Yes?

[diamond.g]

Yes I am referring to what you are thinking. Then perhaps running the best Crysis settings (everything on high) on a Mac Pro is impossible given the Mac Pro doesn't have sli capability (IIRC). Yes?

Basically D3D 10 + 1980x1200 + 32-bit color + All High (or ultra if it exist) + 8xAA+16xAF = 1 hurt GPU.

So yeah you would want more than 2 current GPUs. Now this isn't to say that Nvidias next GPU (I'll call it NV70) will need more than 1 to do the same thing. No one is sure, yet.

[orpheus1120]

I understand everyone's comparing the 8800 card with its PC counterpart lately, but I wander if anyone has any comment on NVIDIA Quadro FX 5600 1.5GB that's offered for our MP, and how it would perform on Crysis?

[diamond.g]

I understand everyone's comparing the 8800 card with its PC counterpart lately, but I wander if anyone has any comment on NVIDIA Quadro FX 5600 1.5GB that's offered for our MP, and how it would perform on Crysis?
Not very well. The Quadro cards have certain settings turned on that make them not so great gaming cards.

[Brianna]

Not very well. The Quadro cards have certain settings turned on that make them not so great gaming cards.

But they usually outperform the other cards offered in the Mac Pro. The 4500 was tested using games and it's statistics were at the Apple site, and it was the fastest gaming card they had at the time. I think it was going up against the Nvidia 6800 Ultra or something like that?

[diamond.g]

But they usually outperform the other cards offered in the Mac Pro. The 4500 was tested using games and it's statistics were at the Apple site, and it was the fastest gaming card they had at the time. I think it was going up against the Nvidia 6800 Ultra or something like that?

Well in this case we all know the G80 isn't too much faster than the G92 at sub 1600x1200 resolutions. Coupled with peoples lack of using AA and AF makes the Quadro a poor choice.

[slackpacker]

Mac Pro 2.8Ghz 2GB,

ATI 2600XT

2/18/2008 10:09:13 PM - XP
Beginning Run #1 on Map-harbor, Demo-Assault_Harbor
DX9 1024x768, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=1, Time Of Day= 5
Global Game Quality: Custom
Custom Quality Values:
VolumetricEffects=Low
Texture=Medium
ObjectDetail=High
Sound=High
Shadows=Low
Water=Low
Physics=High
Particles=Medium
Shading=Low
PostProcessing=Low
GameEffects=Medium
==============================================================
TimeDemo Play Started , (Total Frames: 4100, Recorded Time: 132.23s)
!TimeDemo Run 0 Finished.
Play Time: 169.46s, Average FPS: 24.19
Min FPS: 15.90 at frame 2904, Max FPS: 39.28 at frame 3213
Average Tri/Sec: 2378351, Tri/Frame: 98302
Recorded/Played Tris ratio: 0.14
TimeDemo Play Ended, (1 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/18/2008 10:09:13 PM - XP

Run #1- DX9 1024x768 AA=No AA, 32 bit test, Quality: Custom ~~ Last Average FPS: 24.19


nVidia 8800GT


2/19/2008 9:12:05 PM - XP
Beginning Run #1 on Map-harbor, Demo-Assault_Harbor
DX9 1024x768, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=1, Time Of Day= 5
Global Game Quality: Custom
Custom Quality Values:
VolumetricEffects=Low
Texture=Medium
ObjectDetail=High
Sound=High
Shadows=Low
Water=Low
Physics=High
Particles=Medium
Shading=Low
PostProcessing=Low
GameEffects=Medium
==============================================================
TimeDemo Play Started , (Total Frames: 4100, Recorded Time: 132.23s)
!TimeDemo Run 0 Finished.
Play Time: 106.08s, Average FPS: 38.65
Min FPS: 22.57 at frame 2630, Max FPS: 70.49 at frame 2049
Average Tri/Sec: 3601087, Tri/Frame: 93174
Recorded/Played Tris ratio: 0.15
TimeDemo Play Ended, (1 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/19/2008 9:12:05 PM - XP

Run #1- DX9 1024x768 AA=No AA, 32 bit test, Quality: Custom ~~ Last Average FPS: 38.65

BIIIIG Difference when Playing.

[Nothingness]

Looks like the performance varies a lot on almost the same system. I played the demo of Crysis (i even played it on my 32bit macbook pro ;-) at lowest settings) and loved it. When i clicked the auto detect settings, it put all to very high. I tried this at 1440x900 and AA 2x. The framerate was a bit too slow so moved all except 2 settings to high and then the FSP were very acceptable. Now i will check again and see how many frames exactly.
To be honest, i can understand having 2x AA is nice, but 8x and 16x is somply insanity. Having a higher res and lower AA might even be better than lower Res and high AA. When i do 3d on maya i know that AA 4x is simply a killer and not even worth the effort.

And about the FB-Dimm's : They are basically slower, but the speed on all aspects are almost the same. No matter if it is random read/write of sustained read/write it stays the same. DDR on the other hand can have a huge difference in speed.
Another nice point is that FB-Dimm's use more Watt and get hotter. Because of the memory controller one dimm can hold a theoretical capacity of 16 gig. And with the controller a motherboard can hold up to 32 slots. I'm also not sure if it will have an effect on the game, but i have 8gig ram in quad-channel which is faster then dual-channel.

[Nothingness]

just an additional note for those who don't understand the Anti-Aliasing (AA) =

AA 2x means that your game resolution is 4 times the resolution of your setting, and then scalled down back to the original resolution. 2x means 2 times the width and 2 times the hight.

AA 4x means the resolution that needs to be rendered by the GPU is 16 times as big as the setting's resolution.

I'm not sure if the games and the GPU handles it this way, but this is the theoretical lay-out of AA.

So for those who wanna see if it really matters: take a screenshot of a scene, and change the settings to a higher AA and do another screen. Also have a look at the FPS and then compare. You'll then have a better overlook at what you think is the best setting.

[ildondeigiocchi]

JUST GOT MAC PRO !!!!! GONNA BE PLAYING CRYSIS tomorrow!!! woohoo:):apple::)

[slackpacker]

ALL HIGH with 2x AA nVidia 8800 win xp/ Modded CVARS

2/19/2008 9:38:03 PM - XP
Beginning Run #1 on Map-harbor, Demo-Assault_Harbor
DX9 1024x768, AA=2x, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=4, Time Of Day= 5
Global Game Quality: High
==============================================================
TimeDemo Play Started , (Total Frames: 4100, Recorded Time: 132.23s)
!TimeDemo Run 0 Finished.
Play Time: 159.22s, Average FPS: 25.75
Min FPS: 16.58 at frame 2899, Max FPS: 36.92 at frame 1311
Average Tri/Sec: 12120967, Tri/Frame: 470699
Recorded/Played Tris ratio: 0.03
!TimeDemo Run 1 Finished.
Play Time: 149.34s, Average FPS: 27.45
Min FPS: 16.58 at frame 2899, Max FPS: 37.95 at frame 1956
Average Tri/Sec: 12328493, Tri/Frame: 449071
Recorded/Played Tris ratio: 0.03
!TimeDemo Run 2 Finished.
Play Time: 146.94s, Average FPS: 27.90
Min FPS: 16.58 at frame 2899, Max FPS: 37.95 at frame 1956
Average Tri/Sec: 12433897, Tri/Frame: 445627
Recorded/Played Tris ratio: 0.03
!TimeDemo Run 3 Finished.
Play Time: 146.70s, Average FPS: 27.95
Min FPS: 16.58 at frame 2899, Max FPS: 38.72 at frame 1959
Average Tri/Sec: 12458857, Tri/Frame: 445794
Recorded/Played Tris ratio: 0.03
TimeDemo Play Ended, (4 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

2/19/2008 9:38:03 PM - XP

Run #1- DX9 1024x768 AA=2x, 32 bit test, Quality: High ~~ Overall Average FPS: 27.766666666666666666666666667

[ray_uk]

Crysis's less then stellar performance is due to the fact that it is meant to be playable on current system, but that it should also be able to deliver new levels of detail as new graphic hardware becomes available, so unfortunately to get a solid 60FPS at a resolution like 1920x1200 it becomes necessary to run SLI'd 8800 series graphic cards or Crossfire'd ATI's, the best trade off for a single current card in detail and performance is probably going to be to go for the highest resolution possible where the game still runs smoothly without any AA but with 16xAF.

Its probably not going to be until the Geforce 9 series of cards comes out that a single card will then be able to deliver this level of performance.

[slackpacker]

Crysis's less then stellar performance is due to the fact that it is meant to be playable on current system, but that it should also be able to deliver new levels of detail as new graphic hardware becomes available, so unfortunately to get a solid 60FPS at a resolution like 1920x1200 it becomes necessary to run SLI'd 8800 series graphic cards or Crossfire'd ATI's, the best trade off for a single current card in detail and performance is probably going to be to go for the highest resolution possible where the game still runs smoothly without any AA but with 16xAF.

Its probably not going to be until the Geforce 9 series of cards comes out that a single card will then be able to deliver this level of performance.

You can't equate Crysis's performance with other games like HL2 or Quake.... its just not in the same league graphically. I'm sort of sick with this. There is so much more going on. So Crysis at the Framerates that I am posting is PLENTY Awesome.

[diamond.g]

just an additional note for those who don't understand the Anti-Aliasing (AA) =

AA 2x means that your game resolution is 4 times the resolution of your setting, and then scalled down back to the original resolution. 2x means 2 times the width and 2 times the hight.

AA 4x means the resolution that needs to be rendered by the GPU is 16 times as big as the setting's resolution.

I'm not sure if the games and the GPU handles it this way, but this is the theoretical lay-out of AA.

So for those who wanna see if it really matters: take a screenshot of a scene, and change the settings to a higher AA and do another screen. Also have a look at the FPS and then compare. You'll then have a better overlook at what you think is the best setting.
Actually what you talk about is actually SSAA (Super Sample AA) which isn't really used anymore. MSAA is the norm now. That doesn't require a scene to be rendered at obscene resolutions. Beyond3D post that sorta explains (http://forum.beyond3d.com/showthread.php?t=20359) AA, the post is old so if you resurrect it it will be pointed out.

I also note these test are at 1024x768. At that resolution you are more CPU bound than you are GPU bound, especially if AA and AF isn't cranked. IIRC, 1280x1024 or 1600x1200 is where you start getting more GPU bound.

[Siron]

I just played Crysis for the first time today on my MP with the 3.0 quad and 8800GT. I asked it to autodetect my setting and it turned everything up to HIGH. I checked the AA2 box and ran the game in 1600x1200. It worked so good I couldn't believe it. No stuttering or slowdowns of any sort. I am mightily impressed with this machine for gaming under XP in Bootcam. I'm going to turn settings to VERY HIGH to see if it can handle it.
Alan

[Brianna]

I just played Crysis for the first time today on my MP with the 3.0 quad and 8800GT. I asked it to autodetect my setting and it turned everything up to HIGH. I checked the AA2 box and ran the game in 1600x1200. It worked so good I couldn't believe it. No stuttering or slowdowns of any sort. I am mightily impressed with this machine for gaming under XP in Bootcam. I'm going to turn settings to VERY HIGH to see if it can handle it.
Alan

Same here only I have a 3.2GHz, and I had it on 1680x1050 because that is what my 20" ACD's max settings are. It looks awesome, plays awesome! The water was really cool. I probably didn't need to order that other OC'd 8800GT. Switching back from Crysis in XP Boot Camp to the Mac side with OS X looked the resolution just went down or something. That is how crisp it was. I am amazed by the level of detail in that game. I was chopping trees down with my machine gun, and watching the splinters fly off in random patterns. It's incredible. I also installed the Crysis patch 1.1 before I started playing so I have no idea what it would have been like before it. I can't believe other PC gamers are having trouble playing this game. It looked fine to me. I'll have to install that Crysis performance app tomorrow and play it a little more. I do need a gaming keyboard or something because this little dinky Mac Pro keyboard is weird. But I'm really used to playing games on the PS3 now with a Playstation3 fragFX Controller, and that thing is really cool with Unreal Tournament 3. :)

[slackpacker]

Same here only I have a 3.2GHz, and I had it on 1680x1050 because that is what my 20" ACD's max settings are. It looks awesome, plays awesome! The water was really cool. I probably didn't need to order that other OC'd 8800GT. Switching back from Crysis in XP Boot Camp to the Mac side with OS X looked the resolution just went down or something. That is how crisp it was. I am amazed by the level of detail in that game. I was chopping trees down with my machine gun, and watching the splinters fly off in random patterns. It's incredible. I also installed the Crysis patch 1.1 before I started playing so I have no idea what it would have been like before it. I can't believe other PC gamers are having trouble playing this game. It looked fine to me. I'll have to install that Crysis performance app tomorrow and play it a little more. I do need a gaming keyboard or something because this little dinky Mac Pro keyboard is weird. But I'm really used to playing games on the PS3 now with a Playstation3 fragFX Controller, and that thing is really cool with Unreal Tournament 3. :)

Purchase an Xbox 360 wireless adapter and play with the Best Game controller made the Xbox 360 controller. Its supported in the game directly. It maps out all the buttons automagically.

[slackpacker]

I would really like to know where it says that FB Dimms are slow, ALL my searches say that FB Dimms are fast. In fact since they run series and not in parallel they are much faster than DDR 2, also FB Dimms are DDR2 with other stuff. I just want to understand a little better, I'm not arguing. Thanks

I would like to see some benchmarks.... with all this talk about memory speed. All in all this talk is a waste of time I really doubt it will change the game at all and give any framerate difference at all.

[Brianna]

Purchase an Xbox 360 wireless adapter and play with the Best Game controller made the Xbox 360 controller. Its supported in the game directly. It maps out all the buttons automagically.

Is that the regular 360 controller? I like the mouse and wand w/ the FragFX (http://gizmodo.com/gadgets/peripherals/splitfish-fragfx-mouse-controller-for-ps3-245810.php). It's more like a streamlined PC setup. It's totally sick!

[slackpacker]

Is that the regular 360 controller? I like the mouse and wand w/ the FragFX (http://gizmodo.com/gadgets/peripherals/splitfish-fragfx-mouse-controller-for-ps3-245810.php). It's more like a streamlined PC setup. It's totally sick!

I have the controller... but since I started with K/Mouse I can't play the game with it. Yes its the wired or the wireless XBOX 360 controller. It also lets you use the Wireless headset from XBOX 360 to talk in multiplayer

The best thing is that it is supported right in the game... so no mapping button issues. It just works.

[diamond.g]

I can't believe other PC gamers are having trouble playing this game. It looked fine to me. I'll have to install that Crysis performance app tomorrow and play it a little more.
The reason behind PC Gamers problems is D3D10. Basically my understanding is in XP High = Vista's Medium. This is due to XP not having D3D10 access. There is a mod that allows you to play with the higher detail level, but you would have to google for it.

[ildondeigiocchi]

:apple:I am at the RECOVERY LEVELin Crysis. My Mac Pro has 2.8GHZ (8core) and 8800. All my settings are on HIGH and I have AF to 16X and AA to 4X on 1600 by 1200. I get between 17 and 21 FPS. Very playable and looks stunning! Macs make great gaming PCs especially the MAC PRO. I love Apple!!!!:):apple::)

[bartdesign]

I've been benching crysis 1.1 on my Mac pro yesterday.

I'm using an eVGA flashed 8800 GT, overclocked to 715/1060. With everything high and no aa @ 1680x1050 i get a very acceptable framerate between 35 and 40 frames. When i turn 2xAA on i get between 17 and 28 frames, which makes it laggy IMHO. Not bad.... To bad there aren't any 8800 GT intel SLI drivers, otherwise i would have bought two GT's. Crysis is cool and all, but running UT3 at 2560x1600 with everything maxed out still gives me 60+ fps. That's a pretty good feat aswell.

With the ATI I can't run crysis properly because of the 1024x768 image quality on a 30" ACD. It just makes it to blurry. Also the UT3 performance of the ATI is cryable ;)

BTW: if your intrested in gaming performance on a 4 vs 8 core. I don't think you will notice any difference in terms of frame rates if you choose for a 4 core MP. The most games make use of 1 or 2 cores. Even crysis and UT3.

[diamond.g]

I would really like to know where it says that FB Dimms are slow, ALL my searches say that FB Dimms are fast. In fact since they run series and not in parallel they are much faster than DDR 2, also FB Dimms are DDR2 with other stuff. I just want to understand a little better, I'm not arguing. Thanks

I would like to see some benchmarks.... with all this talk about memory speed. All in all this talk is a waste of time I really doubt it will change the game at all and give any framerate difference at all.

Here is an overview done by Mac on Intel (http://www.maconintel.com/news.php?article=209). They link to some benchmark tests Anandtech did EDIT: look at page 11. The moral of the story Mac Pros run best with 4 FB-DIMMs populated. No more no less.

[slackpacker]

Here is an overview done by Mac on Intel (http://www.maconintel.com/news.php?article=209). They link to some benchmark tests Anandtech did EDIT: look at page 11. The moral of the story Mac Pros run best with 4 FB-DIMMs populated. No more no less.

Article is from 2006 and it still does not have any tests... just guessing. I agree with you that populating all four will give you the best speed. But what I'm saying is what are the REAL performance gains. I would like to see a Crysis before and after test. I bet you don't see a thing as far as performance gain.

[diamond.g]

Article is from 2006 and it still does not have any tests... just guessing. I agree with you that populating all four will give you the best speed. But what I'm saying is what are the REAL performance gains. I would like to see a Crysis before and after test. I bet you don't see a thing as far as performance gain.

This is the best I could find (http://enthusiast.hardocp.com/article.html?art=MTQ1OCw2LCxoZW50aHVzaWFzdA==). Sadly they are only testing Skulltrail and in the end don't really recommend it since it doesn't noticable outperform a cheaper system. Now this is in games mind you. Kyle readily admits if you have program that need 8 cores go for it. Really the only way you can get a similar test is to recreate it. Most here will say Crysis runs great (as it should on a $2500 computer) and that is fine. But those odd ball gamers want superior gaming performance for their coin. Sadly the MP doesn't offer that (yet). Simply, it isn't the fastest gaming box for the money.

[slackpacker]

This is the best I could find (http://enthusiast.hardocp.com/article.html?art=MTQ1OCw2LCxoZW50aHVzaWFzdA==). Sadly they are only testing Skulltrail and in the end don't really recommend it since it doesn't noticable outperform a cheaper system. Now this is in games mind you. Kyle readily admits if you have program that need 8 cores go for it. Really the only way you can get a similar test is to recreate it. Most here will say Crysis runs great (as it should on a $2500 computer) and that is fine. But those odd ball gamers want superior gaming performance for their coin. Sadly the MP doesn't offer that (yet). Simply, it isn't the fastest gaming box for the money.

So... The reason I purchased the Mac Was because it could run Games amazingly and it also runs OS X...The Best!

[diamond.g]

So... The reason I purchased the Mac Was because it could run Games amazingly and it also runs OS X...The Best!

Oh, well what was the point of me finding the comparisions?

[slackpacker]

Oh, well what was the point of me finding the comparisions? You won't ever find a PC that will satisfy the requirement of running OS X and games, and still be able to run games faster than a similarly priced PC. (I hope that made sense)

Oh your pathetic your trying to stir up nothing... go Troll somewhere else.

Yes I could but the whole thing is we are mac owners... the MP IS a great Game machine for Windows. It may not be the BEST We're NOT all going to go out and buy windows boxes. Get with it man.

[diamond.g]

Oh your pathetic your trying to stir up nothing... go Troll somewhere else.

Yes I could but the whole thing is we are mac owners... the MP IS a great Game machine for Windows. It may not be the BEST We're NOT all going to go out and buy windows boxes. Get with it man.

Thank you for correcting me. I shall edit my other post now.

[contoursvt]

I dont get why you got upset. You asked for benchmarks with reference to the difference in ram speeds (FB-DIMM) and you were provided a xeon setup with FBDIMM cmparing to normal DDR2. You didnt seem to care looking at the benchmark or dismissed it as being old even though it was not that irrelevant. The poster of the benchmark basically said "whats the point if you're not going to look at the link" and you jumped on him.

Oh your pathetic your trying to stir up nothing... go Troll somewhere else.

Yes I could but the whole thing is we are mac owners... the MP IS a great Game machine for Windows. It may not be the BEST We're NOT all going to go out and buy windows boxes. Get with it man.

[macguyca]

The whole idea that its fully buffered and error correcting. So it has to check over everything before it can be used. Adding lag time.

Can you use crossfire in Windows boot camp ?

[Siron]

I have been playing this with all levels set to HIGH and AF at 16 for about four hours today. I got to the point where you are at the pit with the alien artifact and the whole place goes crazy. You then start the Relic level where you have to get to the river. At this point my system slows to a crawl with about 10FPS. I exited to the main menu where my mouse pointer was back to normal. However when loading the game again - same problem. I exited to Windows desktop and shut my system down. When I opened the case the graphic card was warm but not hot.
Has anyone else experienced the same problem. I haven't fired it back up yet to see if it was an overheating problem - the exterior of the case was warm but not unusually so and I didn't here the fans - so I'm not sure if they were running at full speed. What controls the fan speeds - is it built into the MB os part of the OS?
Alan

[diamond.g]

Can you use crossfire in Windows boot camp ?

On the new MP, yes you can.

[Jonny427]

On the new MP, yes you can.

Wait, I thought you can't...

[diamond.g]

Wait, I thought you can't...

All ATIs Crossfire needs to work is intercard writes enabled (intercard writes isn't the correct term, but it means basically that) plus at least 2 2X00HD/3X00HD series cards. Nvidias SLi needs (a blessing from Nvidia and) some Nvidia hardware.
So as long as ATI's drivers has the option (which I am pretty sure they do) you should be good to go.

[aaronw1986]

Are you using headphones to play Crysis, or speakers? I have some 5.1 speakers I would like to use, but heard bootcamp does not support surround sound currently? Would installing a sound card (i.e. creative) to use just in Windows be an option?

[Siron]

Are you using headphones to play Crysis, or speakers? I have some 5.1 speakers I would like to use, but heard bootcamp does not support surround sound currently? Would installing a sound card (i.e. creative) to use just in Windows be an option?
I am using a 5.1 surround sound system from Logitech connected via optical and it works great under Bootcamp. I believe that if you insert the Leopard DVD after installing Windows under Bootcamp it will install the necessary audio drivers (Realtek I think).
Alan

[Siron]

Yes I am referring to what you are thinking. Then perhaps running the best Crysis settings (everything on high) on a Mac Pro is impossible given the Mac Pro doesn't have sli capability (IIRC). Yes?

If you have an 8800GT and do an "autodetect" Crysis will set all of the options to HIGH. At this setting it runs great. I do!
Alan

[aaronw1986]

I am using a 5.1 surround sound system from Logitech connected via optical and it works great under Bootcamp. I believe that if you insert the Leopard DVD after installing Windows under Bootcamp it will install the necessary audio drivers (Realtek I think).
Alan

Really? This would be great. And you can actually hear things from behind you and such? Just wanted to make sure, cause there is a thread with people saying you can't do 5.1 in bootcamp.

[Siron]

Really? This would be great. And you can actually hear things from behind you and such? Just wanted to make sure, cause there is a thread with people saying you can't do 5.1 in bootcamp.

5.1 works great (I did have problems until I realized I had the Tosslink cable connected to the "Optical In" port instead of the "Out" port :)). I get sound from my rear speakers but have not checked to see if it's truly the surround effects (to busy immersing myself in the game).
Alan

[aaronw1986]

5.1 works great (I did have problems until I realized I had the Tosslink cable connected to the "Optical In" port instead of the "Out" port :)). I get sound from my rear speakers but have not checked to see if it's truly the surround effects (to busy immersing myself in the game).
Alan

Cool. Which version of Windows are you running?

[Siron]

I checked my speaker setting in XP under Bootcamp and it only shows 2 speakers - not true 5.1 even though it sounds good.
Alan

[aaronw1986]

I checked my speaker setting in XP under Bootcamp and it only shows 2 speakers - not true 5.1 even though it sounds good.
Alan

:(

[OnePumpChump]

I ran Cryisis for the first time yesterday w/ Vista Ultimate 64bit and my Z5500 5.1 surround sound w/ the optical cable and I was able to have surround sound, also true with BF2142 and Oblivion.