Ripple effect affecting the effective buying....

[apple4life]

Yes, yes. We want the Ripple confirmation.

My girlfriend flew to Honolulu today (we live on the big island) for a conference tomorrow and friday and it didn't take her long to get to the ala moana shopping center, the location of the only apple store.... so far from any continental land mass.... and she spoke with an employee that said.....
yes, the ripple effect is supported....

yes, we have ripple effect.... I repeat yes we have ripple effect....




Buy them iBooks!

Get them while they're still hot and have that crisp fresh electrical hardware smell.... ummmmm. TASTY!!!!

 

[AmigoMac]

Ripped effect?? I will update to Tiger and won't cry because I won't see the ripple effect ...

but wait, it looks cool for bragging with Peeceers ...

 

[SpaceMagic]

Thing is, although its trivial in itself, the ripple effect means so much. It means you own a machine which fully supports all the features Tiger wants it to. Although it doesn't matter in the running of the machine in terms of applications you can run, the ripple effect does generally mean you have a faster machine - as some work is offloaded onto the GPU by CoreImage.

It's not necessary but no one likes forking out a grand for something which isn't even up-to-date with the current technologies. The new iBooks are all CoreImage compatible

 

[rdowns]

yes, we have ripple effect.... I repeat yes we have ripple effect....

Most annoying feature....EVER!!

 

[gekko513]

Most annoying feature....EVER!!

Doesn't even look that impressive in my opinion. I don't get the impression of a smooth transition, it's more like a plop the widget in, then start to distort the screen.

 

[joecool85]

I haven't even really seen it yet...but as was said earlier, its not so much that you need the effect, but that if you can use it that means that your system is fully Tiger compliant. To me thats important. I won't install Tiger on my rev a PB for that reason. Well, that and I'll be getting a new mac soon anyway and that will have tiger on it.

 

[Mitthrawnuruodo]

Also: We know the ripple effect is supported. The 9550 was added to Apples Core Image page yesterday.

But thanks for another confirmation...

 

[mpw]

...we live on the big island...

You lucky, lucky ****er.

 

[CoMpX]

I'm so glad they have the ripple effect!! The ripple is really cool and all and it's a really nice touch, but it's not about the ripple. It's about the fact that when you see the ripple, you know that your machine supports CoreImage, and that's important to me.

 

[TheSisko]

I'm so glad they have the ripple effect!! The ripple is really cool and all and it's a really nice touch, but it's not about the ripple. It's about the fact that when you see the ripple, you know that your machine supports CoreImage, and that's important to me.

The 9550 is barely faster than the 9200, I doubt many Core Image effects will get done on the GPU. Apple probably switched because ATI is phasing in 9550 as their new bottom-end GPU.

Just because a card does PS2.0 doesn't mean it's faster/better, compare for example, the FX5200 vs. the GeForce4 Ti cards, which are WAY superior.

For anyone interested in seriously using Core Image, 9550 is NOT enough. The major improvements over the 9200 is the ripple effect and for some OpenGL games that require a fragment programmable GPU for their shader effects (like WoW), though they will still run almost as slow.

 

[Lazyhound]

The desktop version of the 9550 runs 2-3 times faster than the 5200, but I get the impression they down-clocked the mobile version quite a bit. Also, that 32MB kind of kneecaps it.

 

[brepublican]

You lucky, lucky ****er.

Now thats what I'm talkin about!

 

[Lord Blackadder]

LOL, I've been participating in the discussion over the iBook and the Ripple effect, and I'm still running Panther, and will be for awhile.


The 9550 might not be clocked much faster than the 9200, but it has a newer core. That counts for a lot. That, plus ATI's better drivers and the ATI Displays overrides, means it should be superior to the Go5200.

 

[ITASOR]

I think the fast user switching effect is way more brag-capable than the ripple effect.

 

[TheSisko]

LOL, I've been participating in the discussion over the iBook and the Ripple effect, and I'm still running Panther, and will be for awhile.

The 9550 might not be clocked much faster than the 9200, but it has a newer core. That counts for a lot.

In terms of features, it counts for more programmable pixel shaders. In terms of performance increase, it counts for very little.
If you consider the advancements in graphics, almost all is attributed to more pipelines, more shaders, bus width increases and clock speed increases. An optimized design can give you 10-20% if you're lucky, and that's all the 9550 has over the 9200 performance-wise(same bus-width, same clock speed, same number of pipes, same number of texture units, same number of vertex units, same amount of memory).

That, plus ATI's better drivers and the ATI Displays overrides, means it should be superior to the Go5200.

Untrue, the Go5200 in the PowerBook should be WAY superior to the iBook's 9550, since it has double the bus-width (128-bit versus 64-bit). Compare the performance of a 9600SE with a normal 9600 to get an idea.

The desktop version of the 9550 runs 2-3 times faster than the 5200, but I get the impression they down-clocked the mobile version quite a bit. Also, that 32MB kind of kneecaps it.

The ordinary 9550 runs at 250MHz while the iBook runs it at 172MHz. That's a 50% reduction in performance.

Plus the 32MB versions have only 64-bit wide bus to the memory, compared to 128-bit for the 64MB versions. That's a 50% reduction in memory bandwidth, not taking into account the probably lowered memory clock.

 

[Demon Hunter]

LOL, I've been participating in the discussion over the iBook and the Ripple effect, and I'm still running Panther, and will be for awhile.




In terms of features, it counts for more programmable pixel shaders. In terms of performance increase, it counts for very little.
If you consider the advancements in graphics, almost all is attributed to more pipelines, more shaders, bus width increases and clock speed increases. An optimized design can give you 10-20% if you're lucky, and that's all the 9550 has over the 9200 performance-wise(same bus-width, same clock speed, same number of pipes, same number of texture units, same number of vertex units, same amount of memory).



Untrue, the Go5200 in the PowerBook should be WAY superior to the iBook's 9550, since it has double the bus-width (128-bit versus 64-bit). Compare the performance of a 9600SE with a normal 9600 to get an idea.



The ordinary 9550 runs at 250MHz while the iBook runs it at 172MHz. That's a 50% reduction in performance.

Plus the 32MB versions have only 64-bit wide bus to the memory, compared to 128-bit for the 64MB versions. That's a 50% reduction in memory bandwidth, not taking into account the probably lowered memory clock.

That's really interesting... where did you find that info? Do they have that kind of info for the 9700 mobility? I'm interested in what the clock speed is.

 

[iKwick7]

"effect" was used way to many times in the title of this thread

 

[Hattig]

That's really interesting... where did you find that info? Do they have that kind of info for the 9700 mobility? I'm interested in what the clock speed is.

I don't know where he got the information about the graphics clock speed inside the new iBook. Whilst the clock speed of the Radeon 9550 is well documented at 250MHz, and I've seen a mention that a Gateway notebook has a Radeon Mobility 9550 running at 200MHz (300MHz memory), and a review stating the Radeon Mobility 9550 was 100MHz faster than a standard Radeon 9550!

It's a slower Radeon Mobility 9600, based upon the RV350 core, with 4 pipelines and 2 vertex shaders. This core can run up to 350MHz (Radeon Mobility 9600 Pro). I'd be surprised, but then think 'oh, apple!' if they were running it at 172MHz. However the core can accellerate DVD decoding and encoding, which should mean better battery consumption in these tasks. The same core is also used in the radeon mobility 9700, as used in the Powerbook 17". The Radeon Mobility 9200 was based upon the RV280 core.

Regardless, maybe it is time for someone to make a utility to overclock these cards. It seems the mobility 9550's core could go up quite a bit, but its memory might be limited.

 

[TheSisko]

I don't know where he got the information about the graphics clock speed inside the new iBook.

The information is from Apple. They quote the fillrate of the 9550 as 800MP/s, while the fillrate of the Mobility 9600 (at 300MHz) is 1400MP/s. Since fillrate is directly proportional to core clock speed this makes the clock of the 9550 : 800/1400 * 300MHz = 172MHz.

Whilst the clock speed of the Radeon 9550 is well documented at 250MHz, and I've seen a mention that a Gateway notebook has a Radeon Mobility 9550 running at 200MHz (300MHz memory), and a review stating the Radeon Mobility 9550 was 100MHz faster than a standard Radeon 9550!

You are correct about the clock speed of the standard 9550 being 250MHz. Ironically this was also the case with the standard 9200, that is also clocked at 250MHz, and also supposed to sport a faster clock on Mobility versions.

However, Apple clocked the Mobility 9200 to 183MHz.

It seems the mobility 9550's core could go up quite a bit, but its memory might be limited.

I doubt it, since the 9200 should have similar headroom, but barely overclocks anything. I suspect Apple is running their chips at much lower voltage, thus negating any overclockability.

About the memory, I suspect they simply sticked with whatever they had before, since the interface is still 64-bit, and they core clock is about the same.

 

[TheSisko]

That's really interesting... where did you find that info? Do they have that kind of info for the 9700 mobility? I'm interested in what the clock speed is.

I'm pretty sure it's around 390(core)/200(mem) for the PowerBooks. Anyone with a PowerBook can confirm this for you with ATIcceleretor.

 

[amin]

Having just switched from a Rev A 12" PB (no ripple) to a Rev D 15" PB (ripple capable), I don't understand why the ripple is significant.
1) Not all that cool - it was definitely cool the first time I saw Steve Jobs demo it, but it's not as cool as the cube transition in my opinion.
2) So what if it means the system is totally Tiger capable. It doesn't mean that the system will be able to handle everything in 10.5, and my previous PB seemed to handle everything else significant in Tiger just fine.
What am I missing?

 

[Lord Blackadder]

....snip...
Untrue, the Go5200 in the PowerBook should be WAY superior to the iBook's 9550, since it has double the bus-width (128-bit versus 64-bit). Compare the performance of a 9600SE with a normal 9600 to get an idea.

...snip...
The ordinary 9550 runs at 250MHz while the iBook runs it at 172MHz. That's a 50% reduction in performance.

Plus the 32MB versions have only 64-bit wide bus to the memory, compared to 128-bit for the 64MB versions. That's a 50% reduction in memory bandwidth, not taking into account the probably lowered memory clock.

Hum, I thought it was a 128 bit bus; that's too bad.

As far as the downclock, why? The only reason I could think of was thermal concerns. It must exceed their acceptable temp limits when running at "normal" speed. Well, that won't stop people from overclocking it I'm sure.

 

[TheSisko]

Hum, I thought it was a 128 bit bus; that's too bad.

As far as the downclock, why? The only reason I could think of was thermal concerns. It must exceed their acceptable temp limits when running at "normal" speed. Well, that won't stop people from overclocking it I'm sure.

Thermal issues and batterylife no doubt. And it makes sense, since the iBook isn't geared towards gamers, and in normal use, the clock difference means nothing. 64MB, however, would have been nice, specially to avoid Exposé going über-choppy when more than a handful of windows are open.

As overclocking potential goes, will probably be the same as the 9200's, typically around 10%-30%. Why so poor? Because just lowering the clock doesn't decrease the power consumption and thermal generation much, so it's probably accompanied by a heft voltage decrease (why give it more than it needs to maintain the lowered clock?).

Now if only someone came up with a way to voltmod the iBooks...

 

[TheSisko]

Well, it seems as the benchmarks have confirmed that the 9550 only offers modest performance increases.

EDIT: Will be interesting to see if most Core Image effects will be done on the CPU, should be the case, since the PB's FX5200 offers better performance (thanks to the 128-bit bus).