Forums | MacLife

And Ars weighs in.

AMD takes performance crown from NVIDIA with new 4870 X2
ATI's new 4870 X2 dual-GPU / single-card solution launches today, and the card packs all the necessary ingredients to put AMD back in competition for the high-end of the GPU market. R600, it seems, has finally come into its own.

http://arstechnica.com/news.ars/post/20 … 70-x2.html

Bat wrote:

Mr. T volunteers as our [s]guinea pig[/s] tester on the 8800GT front

"Ready to serve." So far, the only ozone from my case came from my $5 450W PSU - That happened about 10min after the card came out of the box.  I've since replaced it with a 750W with lots of current on the +12V rail.  Other than that, I haven't had any trouble as of yet.  (btw, strikethrough doesn't work, here )

Bearfeats wrote:

Though the GeForce GTX 280 is clearly superior in the tests we ran, we think the Radeon HD 4870 is the better choice for Apple's next optional GPU. Why? Because ATI has optimized their Mac Edition drivers and firmware for the Radeon 2600 XT and 3870 so that they run Core Image effects much faster than comparable nVidia cards including the GeForce 8800 GT and Quadro FX 5600.

It *could* be drivers.  rephrase: But I think the problem is somehow related to GLSL, which may or may not be Apple's terf (but likely is).

Bearfeats wrote:

Solution? Make a Mac OS X compatible single card with dual processors like the Radeon HD 4870 X2.

I think that would only happen if there were sufficient demand for an X2 to warrant the requisite driver.  And that won't happen 'till Apple puts slots back in its consumer offerings.  It might also happen if multi-core GPUs become the way of the future (much like multi-core CPUs have become commonplace), but I don't think that will happen, at least not in the short term.  Today's CPUs are multi-core purely out of necessity -- because chip designers have run out of exotic ways to increase single-thread performance.  But GPUs are parallel by nature, so you can always just add more functional units.  OTOH, if ATi could somehow save money by transitioning it's high end offerings to cards consisting of two midrange GPUs with faster RAM, it might be worth it.  I think that's the most plausible way we'll see X2/Crossfire on the Mac. (and it's a stretch)

Bearfeats wrote:

That way the advantages of CrossFire (and SLI)...

SLI?  You'd need more than a driver.

Last edited by Mr. T (2008-08-13 10:37 pm)

Bat wrote:

Mr. T wrote:

Bat wrote:

Mr. T volunteers as our [s]guinea pig[/s] tester on the 8800GT front

(btw, strikethrough doesn't work, here )

http://www.maclife.com/forums/post/1576362#p1576362

Someone's http://homepage.mac.com/oatmeal/MAF/maxes/mrtMax.gif not done his homework. http://homepage.mac.com/oatmeal/MAF/maxes/bounce.gif

I'd have used "ahem" instead, but I can see that you are a man of principle   Hey, maybe if a bunch of us start using [s][/s], M|LF will implement the feature for fear of their forums becoming an ugly mess of tags.  We can do it.  They'll see!!

Bearfeats wrote:

[..]

Apple's terf...

It's 'Barefeats' and 'turf.'

Oh, the irony and agony -- it's killing me (but mostly the agony)!  I mean, it's one thing to misspell turf (even with systemwide spell check), but writing Bear instead of Bare, not once, but three times.... I am absolutely and utterly ashamed of myself.

Ok so with the NVIDIA GeForce 8600M GT with 256 MB & 512 MB cards more or less defective, how much longer can Apple (and everyone else) keep using them in their machines?

Mr. T wrote:

Until the supply runs out?

well yes that would be one option, but why keep using what you will have to just replace?

Yup. And busted replacement parts are all there is.

Nice covert war going on; NV trying to push off their liability with firmware updates that only speed up the fan, and HP etc. extending their warranties to shift it back.

Last edited by Bat (2008-08-16 5:10 pm)

One more TPC! NVIDIA will offer a upgraded GTX 260 in mid-September

The graphic card battle is getting more fierce right now. Our source just inform us that NVIDIA will upgrade GTX 260 in mid September, offering one more TPC inside the G200 GPU.

The original GTX 260 have 8 TPCs (Texture Processing Cluster) totally, but the new GTX 260 have 9 TPCs. That means the numbers of shader processor have been increased from 192 to 216. In G200 chip, one TPC contains 24 shader processors.

Other than the TPC upgrade, the other specs are keeps the same. That means the card still runs at 576/1242/999Mhz (core/shader/memory), Memory interface keeps 448bit, and memory size is 896MB.

We guess that this upgrade will help the card to suppress HD 4870. But HD 4850 X2 is right around the corner…

http://en.expreview.com/2008/08/21/nvid … september/

GTX 260+? One more number to remember...

Heh, guesstimating on the GPU market hasn't been that great of an idea recently. Either Intel is setting a ultra high bar, or they are going to come up wanting.

OTOH, NVision's given tech journos like Anand some face time with Jen-Hsen and David Kirk.

Out of nowhere Jen-Hsun threw out quite possibly one of the most important statements I'd ever heard him make, that NVIDIA's mobile strategy is to "completely focus on smartphones".

Jen-Hsun went on to say NVIDIA believes the "phone will become the next personal computer".

This highlights an important potential strategy for NVIDIA. If Intel/AMD push NVIDIA out of the conventional PC market with their GPUs/platforms/CPUs, NVIDIA needs to look elsewhere: the mobile phone market is an interesting outlet. Remember this slide from IDF?

By 2012 Intel estimates that it will be shipping ~200M desktop CPUs, ~360M notebook CPUs and around 200M Atom CPUs. The mobile CPU market is just as large as the mobile GPU market, the difference being that margins are much lower. But it's quite feasible that NVIDIA could see a good hunk of its revenue coming from that market, should it be forced out of the GPU market. But it's worth noting that the Intel/AMD/NVIDIA GPU war looks like it will take at least 3 - 5 years to see through.
...

A reporter from The Inquirer asked Jen-Hsun to clarify what other GPUs might be affected by the manufacturing issue and to basically get more specific, publicly with what users can expect.
...

The question Jen-Hsun didn't answer is what GPUs in specific that this problem impacts and whether or not it extends to the entire line of G8x GPUs, desktop and mobile.
...

What does NVIDIA think about Lucid and Hydra? Well, some good things and some bad things. Of course no one I asked gave me anywhere near an answer about how they thought SLI would be impacted if Lucid met their goals. And there's really no good way to dodge that either; it always ends in sort of a trailing off into other things attempt to change the subject.
...

So far we haven't been too impressed with NVISION, but access to folks like Jen-Hsun and David Kirk has thus far been worth it.

We're off to hear Epic's Tim Sweeney speak about Unreal Engine 3, check back later for more updates from NVISION.

Smartphones, eh? There's one I've been hearing about lately...

NVISION 08 - Jen-Hsun Talks Larrabee, Mobility, VIA and More

Criminy. Even I'm losing track.

With that, lets take you through, point by point, the Nvidia 6 page PDF that they issued as a PCN. Page 5 is blank, and Page 6 is the standard legal disclaimer, so we will skip those. The PCN is dated May 22, 2008 on the bottom of pages 2-5, July 25 on the bottom of Page 1, and Page 6 is undated. The first big problem is that it is entitled "G86 Desktop Products" with a subtitle " Change Namics 8439-1 Underfill material to Hitachi 3730". Above that there is " Product/Process Change Notice", the usual NDA only disclaimer.

Remember how Nvidia swore up and down that desktop parts were flat out not affected? Remember how we said that all G84 and G86s were because they were the same ASIC? I guess they decided to change this underfill material to better color coordinate with the substrate hues, given the cost of testing, qualification and other work that needs to be done, you certainly wouldn't want to change it for no good reason. The old one worked just fine, right? Not defective either, they said so. Then again, they said the problem was contained to HP as well.

The official Nvidia explanation is that if you change one SKU, you change them all, so this isn't a big deal. Testing and validation costs be damned, you take a 'working' and near EOL part and change it on a whim because they changed another part. OEMs love that, as do stockholders.

Nvidia changes desktop G86 for no reason

Nothing's wrong, but we're fixing it anyway. Because.

HOT ON THE heels of its denials that anything is wrong with the G92 and G94s comes another PCN that shows the G92s and G92b are being changed for no reason. Yup, the problems that are plaguing G84 and G86 are the same that affect seemingly all 65nm and now 55nm Nvidia parts.

This PCN is very similar to the one linked above, and the formatting is almost almost exactly the same, so we won't cover all the details, just the pertinent points. This one is much more important, it confirms that the problems are not confined to the 65nm products. Since Nvidia told us the last one was unimportant and refused to give it to us, we didn't bother asking this time, we just took notes when they were shown to us at a recent conference.

Nvidia 55nm parts are bad too

I am so glad I don't have a recent NV part.

The NVidia News (Labor Day Edition).

NVIDIA BIOS Editor (NiBiTor) version 4.5 is now available.

Why Nvidia's chips are defective
Part One A long and complex story

Why Nvidia's duff chips are due to shoddy engineering
Part Two The underfill

NC's nightmare is in the Rumor Mill.

SPECULATION FROM INDUSTRY insiders has it that Intel is having significant trouble getting publishers and developers to programme for Larrabee.

INQ sources reveal Chipzilla is offering to shell out millions of dollars to anyone who will bravely take the plunge into the great Larrabee unknown, an incentive which still doesn’t seem to be doing the trick, say some.

As with any new architecture, Larrabee is considered a risk, and a potentially costly one at that, scaring developers into uncertainty about whether or not to put the work in.

It seems that the only real way to force the developers’ hands, would be if Intel were to win a contract for a major console, meaning publishers would then have to programme for it.

This means Chipzilla is now under immense pressure to get itself a console deal. And apparently all roads lead to Vole.

And our well-placed if publicity-shy sources claim Intel has offered Microsoft a very sweet deal indeed in exchange for pushing AMD out of the running for Mightysoft’s upcoming Xbox 720.

Intel needs a console deal to get Larrabee off the ground