Early Benchmarks: Dual 1GHz

[AmbitiousLemon]

Re: Re: Ridiculous responses...

Originally posted by deepkid


That is a very troubling statement. It is as scary as some court of law simply glancing at an accused person who "might" look guilty and charging them so without due process.

If the tests that Barefeats performed are suspect, they should be not be held in such high regard. It does not matter what Apple's or Barefeats' opinions are, the results should be based upon sound benchmarking.

the ONLY reason these benchmarks are 'suspect' is that 1) it doesnt say what some of you want it to say, and 2) you are too lazy to go to barefeats.com yourself and read hwo they were performed.

dont get me wrong, i think this is a great update. but its important to have this info because it is contrary to what apple has told us. apple tells us the new 867 and new 1ghz are faster than the old 1ghz. this does not seem to be the case. but more benchmarks are on the way, and we cant test the new 1.25ghz until the thing ships. This is information. it should help people make purchasing decisions. all the info isnt in yet, so hold off on making any drastic decisions. but throwing this out just because you dont like what it says is childish and not in your best interests.

 

[eric_n_dfw]

Originally posted by ImAlwaysRight
I'm eligible for Apple edu discount, so picked up a new dual gig for $2249 shipped (no tax). After paying shipping at a mail order house, it is about the same price as the old dual gig, and I like the newer better.

What mail order house is offering you edu discount? (I haven't found one that will yet)

 

[deepkid]

Re: Re: Re: Ridiculous responses...

Originally posted by AmbitiousLemon


the ONLY reason these benchmarks are 'suspect' is that 1) it doesnt say what some of you want it to say, and 2) you are too lazy to go to barefeats.com yourself and read hwo they were performed.

Was this a direct response to me?

I can't speak for the other forum members, but I looked at the Barefeats data several times -- and its too muddy.

Before getting into the specific tests, he should have given the OS and versions used for all machines...RAM and HD types and speeds. You have to perform this test using Jaguar, because that is the OS of today, not OS 9.2.

That would have exacted more credibility for him.

I personally don't have an emotional tie to the results. I own/use several macs and I'm pretty happy with them. I will replace my older powermac, so its not a matter of wanting the results to verify some pipe dream.

I simply have an interest, like many of us, in receiving the most sound benchmarks possible in order to make a better purchasing decision. It does not matter who performs these, just that they are as accurate as possible.

It appears that we will have to wait a while longer to get them.

Thanks.

 

[carta]

Barefeat's eMac review

I visited the Barefeats site for the first time because they reviewed an eMac and the new iMac, both with 700 MHz processors. In the conclusion, the tester said that both machines were similar, but that he gave the nod to the new iMac because it had a significantly faster hard drive seek time.

While Apple's tech specs are only so informative, the eMac and new iMac appear to have the same 40 GB ultra ATA hard drive.

So the Barefeats conclusion really should be that the machines are equal (given equal hardware) and that they really comparable. But he goes on to bash the eMac's "flimsy" hard shell, suggesting that the machine is less rugged than its LCD cousin.

 

[ImAlwaysRight]

Originally posted by AmbitiousLemon
i think some of you need to actually LOOK at barefeats.com before attacking the benchmarks. you are screaming and yelling about things that are flat out wrong. he does say very specifically how each machine is configured and he also says very specifically how he set up the tests.

Umm, two things. First, he does NOT say how each machine is configured. Where do you see how much RAM is in each system? Where do you see what video card was used? He lists system specs like max RAM, available video cards, but it does not say how much RAM is in each system. LOL - who needs to LOOK. The new dual gig only ships with 256MB RAM, and the old dual gig has 512MB. Assuming he just used stock configs, that RAM amount would make a difference in tests. Second, I just looked again, and now the barefeats website has been updated. I can't control the fact that the guy revised his website after I viewed it and made my remarks.

Originally posted by AmbitiousLemon
his benchmarks in the past have been very reliable, and have been very fair.

Most of his tests seem fine. He's had some questionable test results in the past though. Perhaps search www.xlr8yourmac.com if you want to find discussion on it (that's probably where I was tipped off to it, but its been a while).

Originally posted by AmbitiousLemon
comparing all the dual 1ghz machines is the best way to test whether the new bus and ddr improve performance. if you can not understand basic scientific method then please dont criticize.

If you want to judge the changes in architecture then you need to move the same hard drive and video card from one machine to the next. Have the same amount of RAM installed. This is the only way to give an accurate result. I suppose I can put this in bold, italic, or even 72 point font, yet some people will still fail to understand... Just because you claim to be more intelligent does not make it so.

And remember. ImAlwaysRight (heh-heh)

 

[QuiteSure]

Overcompensating?

Originally posted by yadmonkey
I can honestly say that these new Dells we got with the really fast bus (533 mhz, I think) and super-over-clocked P4s are overcompensating for something.

What does this mean, "overcompensating?"

 

[Timo_Existencia]

Not disputing...

IAMRIGHT...

I agree that these benchmarks are preliminary; I suspect that BareFeats will update as time and information allows.

I'm curious, however, you don't seem to be as scrutinizing of the benchmarks that Apple has put out. Why? For example, I don't think it's fair to put dual processors against single processors. Do you?

BTW, again I'll ask, where is the Old Dual 1 gig being sold for $2199?

 

[FatTony]

Re: Overcompensating?

Originally posted by QuiteSure


What does this mean, "overcompensating?"

o·ver·com·pen·sa·tion n.

Excessive compensation, especially the exertion of effort in excess of that needed to compensate for a physical or psychological characteristic or defect.
Source: The American Heritage® Dictionary of the English Language, Fourth Edition
Copyright © 2000 by Houghton Mifflin Company.
Published by Houghton Mifflin Company. All rights reserved.

 

[Uragon]

Re: Re: Re: Ridiculous responses...

Originally posted by AmbitiousLemon


..... more benchmarks are on the way, and we cant test the new 1.25ghz until the thing ships. This is information. it should help people make purchasing decisions. all the info isnt in yet, so hold off on making any drastic decisions. but throwing this out just because you dont like what it says is childish and not in your best interests.

I agree...this is just a preliminary information.... I'am waiting for other benchmark before deciding whether to buy the new ones...

 

[eirik]

L3 Cache

Does the size of L3 cache, 1 MB versus 2 MB, always make a difference for all performance tests?

I suspect that it doesn't.

Is it not conceivable that 1 MB of L3 cache is more than enough cache for application instruction, static data, frequently repeated data, and intermediate data (?)?

In time, I hope some performance tests can be run such that the testers alter the size of the L3 cache so that they're the same in both machines, affording us a more focused comparison of the architectural changes. Is there some reason that this cannot be done by a skilled individual (other than warranty issues)?

Eirik

 

[Sun Baked]

Overcompensation

The Porsche effect?

 

[ImAlwaysRight]

Re: Not disputing...

Originally posted by Timothy
IAMRIGHT...

Are you trying to get my attention or are you saying that you are always right. Because only one of us can always be right, and that is me. Look at my name. It says so, so you can't dispute that. 🙂

Originally posted by Timothy
I'm curious, however, you don't seem to be as scrutinizing of the benchmarks that Apple has put out. Why? For example, I don't think it's fair to put dual processors against single processors. Do you?

Actually, I'm being sarcastic. The Apple results, although no doubt accurate when tested, are definitely biased. I wonder how many hundreds of tests Apple performed just to put these two graphs I reproduced on their website. For the record, I do NOT believe a dual G4 867 to be faster than a single 2.54MHz Pentium 4, other than maybe 1% of the time.

Originally posted by Timothy
BTW, again I'll ask, where is the Old Dual 1 gig being sold for $2199?

I think I saw it at MacMall.com. Any place that has it in stock should have it at $2199, including your local CompUSA or other. MacWarehouse even had refurb duals for $2099, but for $100, I'd go with the new sealed box if you can still find it.

 

[yadmonkey]

Re: Overcompensating?

Originally posted by QuiteSure


What does this mean, "overcompensating?"

Perhaps I should have simply said "compensating" for accuracy. What it means is that through my very unscientific tests (general usage, including Photoshop), they don't seem to perform much better or worse than the Quicksilver dual-1ghz I use. Therefore, there must be some flaw in the overall architecture or system software (XP in this case) which requires them to boost the bus and clock speeds so high.

BTW, XP ain't half-bad.

 

[soosy]

Man you guys are ridiculous

The criticism of barefeats tests seems a little harsh and unwarranted to me. If the new dual-1Ghz aren't any faster than the older dual-1Ghz's then that is a little dissapointing to me. I wouldn't normally expect old and new machines using the same chip to perform differently, but apple is very much stressing the benefits of the new architecture (which is understandable, they would want to have something more than the chip speed increase to make these machines seem "all new")

The "crippling" someone else was referring to is that the new dual-1Ghz has a smaller L3 cache than the old one. So if barefeats results are accurate, then maybe the new architecture is helping, while the smaller cache is hindering. They could have kept the cache the same, but it probably wouldn't have given as big a performance difference between the new 1Gz and the 1.25Ghz machines. I'm pretty sure they've done similar things before when speedbumping machines, so it's understandable, even if I don't exactly like it. I also don't like it when they "manufacture" limitations to seperate product lines. (for example, preventing external display spanning on the ibook when the graphics card is perfectly capable.)

Anyway, I think the new powermacs are a good step forward, even if lack a certain amount of the "wow" factor. It's certainly great that the high-end dual 1ghz are now in the middle at a significantly cheaper price. More bang for the buck.

I do like knowing the difference the between Apple Hype and reality. It's bothered me lately that they claim to "invent" things that are already out there. (e.g. Not acknowledging O'Meara's G-Force as the inspiration for iTunes visualization. Instead they were like "We asked ourselves what would it be like if you could see music"). Uh, okay, I'm off topic now.

 

[yadmonkey]

Re: Not disputing...

Originally posted by Timothy
BTW, again I'll ask, where is the Old Dual 1 gig being sold for $2199?

oops...

http://www.clubmac.com/clubmac/families/powermac/default.asp?estore=41147000

 

[AmbitiousLemon]

Re: Re: Re: Re: Ridiculous responses...

Originally posted by deepkid


Was this a direct response to me?

I can't speak for the other forum members, but I looked at the Barefeats data several times -- and its too muddy.

Before getting into the specific tests, he should have given the OS and versions used for all machines...RAM and HD types and speeds. You have to perform this test using Jaguar, because that is the OS of today, not OS 9.2.

That would have exacted more credibility for him.

I personally don't have an emotional tie to the results. I own/use several macs and I'm pretty happy with them. I will replace my older powermac, so its not a matter of wanting the results to verify some pipe dream.

I simply have an interest, like many of us, in receiving the most sound benchmarks possible in order to make a better purchasing decision. It does not matter who performs these, just that they are as accurate as possible.

It appears that we will have to wait a while longer to get them.

Thanks.

not a direct response to anyone. just to the general crowd all saying the benchmarks are questionable. he does list the info you mention is missing. since you missed it i assumed you didnt look at the page. he clearly says what os he used and what the specs are on the machine (including RAM etc). he can post all this info but if you refuse to look... not his fault. if you cant read a table...sorry but that makes your intelligence suspect not his tests. not trying to get personal or anything, but i mean the info is there. attacking him based on faulty info isnt fair. try to take more than a cursory look at the data before launching into your attacks. the reason he didnt put the same amount of ram/hard drive etc is because these things are changes that have been made between the models. these are just parts of the thing he is testing. the max ram has been changed, the ata controllers have been changed. we want to see if all this makes a difference.

I suppose I can put this in bold, italic, or even 72 point font, yet some people will still fail to understand... Just because you claim to be more intelligent does not make it so.

wow what great logic. i cans ee why you were unable to understand a simple table showing the specs of the machines.

 

[ImAlwaysRight]

I'll chime in about the use of the term "crippled" associated with the new dual gig. How is it crippled? Maybe because the former mid-range had 2MB L3 cache, while the new mid-range has L3 cache. Yes, if our new mid-range were a 933MHz single processor with 1MB L3 cache, 256MB RAM, and the other specs as the former mid-range 933, then I would call our new mid-range crippled.

But, what did Apple do here? We have a whole new computer here. They gave us a speed increase from 933MHZ to 1GHz. Wait, what? They added a second processor? Since when? Sweet. Now dual processors in the mid-range when there was only one. L3 cache is 1MB. Awww, com'on, the previous mid-range had 2MB L3. But, considering the other improvements, like DDR RAM, ATA/100 controller, 166MHz front side bus, yes, this is a good upgrade. Nothin' crippled here.

THE NEW DUAL GIG IS NOT THE OLD DUAL GIG. IT IS A REDESIGNED COMPUTER. THINK OUT OF THE BOX. This is not a hand-me-down. You've got new clothes, kid. "Aww, but dad, my suit only has two buttons on the front and brother John's had three!" "Dangit, kid, you got the wool instead of polyester, and we had yours tailored while John's doesn't fit quite right. Quit yer bitchin'!"

Or something like that.

It seems in the effort to test the ARCHITECTURE of the new Powermacs, people want to take the two models that have the same MHz speed and compare them. But they are taking the high-end and comparing it with the mid-range. But in reality, I'm not sure this is fair, EVEN BY USING THE SAME HARD DRIVE AND VIDEO CARD BETWEEN SYSTEMS, AND HAVING THE SAME AMOUNT OF RAM, to put the old G4 dual gig vs. the new, because of the L3 cache disparity. To me, this is not a true test of the architecture. We need a dual gig processor with 2MB L3 cache to properly test the architecture of the new system bus, ATA/100, and DDR RAM, don't we?

 

[whatever]

Alright here are my two cents. If you want to compare the old dual 1 GHz PM against the new, then one of the key metrics, which is missing is cost. Like one person mentioned several times, you're comparing a High End machine to a Mid Range machine. So one of the factors which make these machines either High End or Mid are their costs.
The trick here is some way including the price into the equation. And then we can determine the true worth of these machines. Once that fomula is compete run it against the old and new high end machines.

Bottom line is that many of you are hung up on hardware specs, but how many of you are actually using new machines, for the most part older Macs depreciate much better than anything else out there and that's the problem, they just don't die!

 

[Timo_Existencia]

For those interested...

Thanks for the link to ClubMac and the pricing on the Old 1 gig.

It is selling there for $2129. For those who think that the dual 867 on Apple's site is a great deal...if you add ram, larger hard drive and superdrive to match the old 1 gig...you'll get a price of $2099. Only $30 less than the original 1 gig is selling at ClubMac. That's a pretty good deal for the old 1 gig.

🙂

 

[Sun Baked]

Hello McFly, (TAP, TAP) anyone home?

Here's a chart comparing the features and specs of the three Dual 1GHz systems:

CHART/TABLE

Unless my brain is still asleep, this is only a copy of the data from Apple - not an as tested table.

When I compiled my list above it's what I used.

However I neglected to mention that 2 of machines should boot into OS 9 - which should give better access to hardware than OS X - and there are still people who need to boot OS 9, so some OS 9 testing may be of interest to people.

 

[Frobozz]

Still a VERY relevant observation

No matter how taxing these tests were, they were test used to show real world usage. Most users will not be doing Cinema4D renders for 24 hours straight, even though I may do things like this.

The point is that, for the average user, these systems are not faster. They are an excellent glimpse of what is to come-- either the 7470's or the Power4. The architecture improvements are very welcome, but my 2 x 1 GHz Mac from February is lookin' like it was a great investment.

Apple's gotta get a new PowerPC processor-- but otherwise it's doing just fine.

 

[porovaara]

Re: Re: Re: Re: Re: Ridiculous responses...

Originally posted by AmbitiousLemon

saying the benchmarks are questionable. he does list the info you mention is missing. since you missed it i assumed you didnt look at the page. he clearly says what os he used and what the specs are on the machine (including RAM etc).

Where?

 

[soosy]

THE NEW DUAL GIG IS NOT THE OLD DUAL GIG. IT IS A REDESIGNED COMPUTER. THINK OUT OF THE BOX.

Yes it is a new system, which is why I would expect it to perform better than the old systems. 🙂

...people want to take the two models that have the same MHz speed and compare them

Yes that's exactly what we are doing. I guess we just have a difference of opinion here. I think it is perfectly reasonable to compare the two.

And like I said before, the new model is great if only for the price reduction. 🙂

 

[ImAlwaysRight]

Re: Re: Re: Re: Re: Ridiculous responses...

Originally posted by AmbitiousLemon
he does list the info you mention is missing. since you missed it i assumed you didnt look at the page. he clearly says what os he used and what the specs are on the machine (including RAM etc). he can post all this info but if you refuse to look... not his fault. if you cant read a table...sorry but that makes your intelligence suspect not his tests.

Let's talk about looking at pages for a minute. In fact, go back to the barefeats site. It says: "Here's a chart comparing the features and specs of the three Dual 1GHz systems:" He then goes on to give Maximum Memory, Standard Graphics Card, and Optional Graphics.

So do you believe he had each system maxed out for memory? Which video card was in the test systems, the standard card, or the optional? Rob Morgan is simply stating the specifications of the machine, he's not telling you how his test systems are configured.

EDIT:OK, I'll be nice. I won't say anything about intelligence.

 

[kenohki]

Re: Re: Re: Re: Re: Re: Ridiculous responses...

Originally posted by ImAlwaysRight

Hah! Let's talk about looking at pages for a minute. In fact, go back to the barefeats site. It says: "Here's a chart comparing the features and specs of the three Dual 1GHz systems:" He then goes on to give Maximum Memory, Standard Graphics Card, and Optional Graphics.

So do you believe he had each system maxed out for memory? Which video card was in the test systems, the standard card, or the optional? Rob Morgan is simply stating the specifications of the machine, he's not telling you how his test systems are configured.

I can see why you misunderstand what I have to say when you can't even get this right.

So, please look, think, and then write, in that order.

And remember. ImAlwaysRight! (woooo-hoooooo!) (can you tell I play RTCW?)

Yeah, those are just the max specs. He doesn't list configuration on any of the machines tested...not even under the "How I Test" page. You can take these benchmarks with a grain of salt.