Don't get fooled by .... nvidia attacking Ati

[jvd]

I dunno dave , in the slides they talk about the 9550 , the 9250 , native pci-e and other general things across the product lines.

Not only that but ati was addressing the speed increase in doom3 performance with the newest drivers in that slide comparing two equaly priced cards.

I'm asuming this is the slide in question http://www.hardocp.com/image.html?image=MTA5NjQxMjE3OHBPNUdCc2FrRG9fMV8xMV9sLmdpZg==

 

[Dave Baumann]

The first slide addresses some general points and the rest of the presentation is specifically about X700.

 

[jvd]

The first slide addresses some general points and the rest of the presentation is specifically about X700.

I dunno , i'm just saying i don't see a problem. Its not like they are comparing the x800xt pe to a 6600gt thats using drivers from 3 months ago

 

[Fodder]

Just spotted a weighty article on the ATI PCIE compliancy deal from Charlie at TheInq. I'm not sure on the accuracy of it all, but it makes for an interesting read.

 

[radar1200gs]

LOL! So much for the much vaunted (in these forums) "superiority" of ATi's on die PCI-Express bridge.

It would appear it has caused ATi at least one respin of their chips and still can't acheive compliance.

Meanwhile, nVidia's "stupid", "expensive", etc, etc external bridge chip gets the job done properly with compliant products in the marketplace.

I wonder if the Low-K process is affecting ATi's ability to finetune the signal parameters

 

[Bouncing Zabaglione Bros.]

Just spotted a weighty article on the ATI PCIE compliancy deal from Charlie at TheInq. I'm not sure on the accuracy of it all, but it makes for an interesting read.

The last article he wrote on ATI was complete tosh, so I'm not treating this one as gospel.

 

[Dave Baumann]

Personally I'd wait to hear something more definite that Charlies report - when the document that was being circulated by gainward was passed to me I asked someone at Dell what they thought and they directed me back to the SIG statement.

As for a respin, this is fairly common for multiples of reasons - even if some of these claims are true you don't know that the earlier revs ever made it into systems.

As for the low-k comment, this makes no sense - X600 is low-k and has been compliant for some time.

 

[digitalwanderer]

The last article he wrote on ATI was complete tosh, so I'm not treating this one as gospel.

OMG, I missed that connection...isn't he the guy who wrote that piece explaining why nVidia was kicking ATi in the OEM market when the reality was the exact opposite?

I'm taking anything he writes with a rather LARGE grain of salt and waiting for confirmation, I wish I could post up the e-mail exchange I had with him after his last "article".

 

[digitalwanderer]

Just spotted a weighty article on the ATI PCIE compliancy deal from Charlie at TheInq. I'm not sure on the accuracy of it all, but it makes for an interesting read.

I saw this Inquirererer story about ATi trying to get the PCI standards loosened, but when I saw it was by Charlie Demerjian I remembered his last article and remembered how full of shit it was...so I dropped a line to me friend Patti at ATi who just responded to me:

The Inquirer appears to have based their article on the same (incorrect) information that is being circulated in the form of an anonymous "whitepaper". The document (and the Inquirer article) claims that a set of "eye diagrams" show that ATI's PCI Express cards do not meet the standards for electrical compliance. Because the circumstances of the testing are unknown, we will not comment on the actual test results except to say that they are not valid.

The anonymous document being circulated is inaccurate to say the least. Unfortunately the Inquirer has chosen to base their article on that incorrect information and therefore the conclusions they draw are also incorrect.

Several products, including the RADEON X300, were submitted for PCI-SIG compliance testing on September 13 and 14 during the Milpitas Interoperability Plugfest. The results of the testing are not expected for several weeks. However, it's worth noting that we have completed stringent interoperability and compatibility testing with our customers, including all of the major OEMs, and to date we have shipped millions of PCI Express products.

For more information on this document please see the PCI Express(tm) Link Quality Notice here http://www.pcisig.com/developers/compliance_program/quality_notice.

My thanks to Patti for clearing this up so quickly for me. 8)

Linky-poo.

 

[Geo]

ok, I don't care if Nvidia and Ati wants to sling mud and lie about each others products, but now Ati has gone too far!

Well, at least they cited in the way you are supposed to do these things, without taking b3d's logo or look 'n feel, and providing the actual url for interested parties to go read for themselves exactly what was said and in what context.

 

[flick556]

Having worked for an unnamed retailer I know why these documents always get leaked. During the developer fairs and other such events, companies show certain people this kind of stuff in hopes it will trickle down into the way salesmen interact with customers. It’s usually not a big official presentation or anything but if you ask they have these on hand. It’s basically what their pr people are supposed to be telling us but they do show the actual slides sometimes.

My point is these are not top secret or highly guarded documents, they just are not “supposedâ€￾ to be freely available to the public. I'm sure review sites can get a copy if thier good.

 

[Pete]

At least, however, the the pipeline diagram is still in there.

It's been pointed out that you seem to have dropped the fragment crossbar from your NV43 diagram. Slip of the mouse in MS Paint?

 

[Dave Baumann]

Well, a cross bar usually indicates that more than one input can go to more than one output (as in the case of NV40), however with everything set up in "quads" there is only one output in NV43, which, IMO, negates the need for an actual crossbar. I'd guess its more than likely just a FIFO there.

 

[radar1200gs]

Personally I'd wait to hear something more definite that Charlies report - when the document that was being circulated by gainward was passed to me I asked someone at Dell what they thought and they directed me back to the SIG statement.

As for a respin, this is fairly common for multiples of reasons - even if some of these claims are true you don't know that the earlier revs ever made it into systems.

As for the low-k comment, this makes no sense - X600 is low-k and has been compliant for some time.

Desperately awaiting a loosened up PCI Express Spec are we Dave? If the cad can't meet the specs, the specs will just have to meet the card...

Like the Inquirer points out it will be fascinating to see how various first gen PCI Express cards perform on future chipsets when they are no longer the latest and greatest cards. I have no doubt about who I think will have no problems running flawlessly and who will run into trouble and require all kinds of workarounds.

Of course you could have pretty much predicted this sort of outcome just by watching how ATi and nVidia coped with the introduction of AGP 8X...

 

[martrox]

Jesus, Radar....... You only show up here to offend people and say very unintellegent things.... Maybe you should run for political office!

 

[radar1200gs]

Jesus, Radar....... You only show up here to offend people and say very unintellegent things.... Maybe you should run for political office!

Personal attacks aren't supposed to be made in these forums. I thought that you, as a moderator would be the last person needing to be reminded of that...

 

[Dave Baumann]

Desperately awaiting a loosened up PCI Express Spec are we Dave? If the cad can't meet the specs, the specs will just have to meet the card...

No, I saying that Charlies past post don't have much basis in reality - his excuse is "everything I wrote is true as far as I know it", however his research is particularly lacking hence his claims are were false.

However, I was pointing out that your low-k comment was particularly out of place - it was also the case that Tech Report used X600 in their pre-release VIA PCI Express chipset test.

 

[DarN]

Jesus, Radar....... You only show up here to offend people and say very unintellegent things.... Maybe you should run for political office!

Just look at the amount of sigs he's featured in. Even yours...

 

[radar1200gs]

Just because a circuit performs well on one die doesn't mean it will perform well on all dies, especially where low-k and transistor count is concerned.

X600 has far fewer transistors than its newer cousins.

 

[Dave Baumann]

Just because a circuit performs well on one die doesn't mean it will perform well on all dies, especially where low-k and transistor count is concerned.

X600 has far fewer transistors than its newer cousins.

Again, RV380 (X600) has already passed the tests and this is a low-k core. RV380 is also the reference platform for ATI's PCIe. RV370 (one of X800's "newer cousins") is also exactly the same transistor count as RV380.