OFFICIAL WORD FROM NVIDIA is that gaming is now a footnote for the formerly leading graphics company, its latest and greatest Fermi chip won’t see the light of day until at least Q1 of 2010, and Fermi’s clock speeds so far are massively off what the firm had hoped for. Yup, it is a mess, just like we told you.
If you read the link here, the boys in green are spinning hard, with a big show starting today, but have nothing to show. Sure, Nvidia will have one or two ‘puppies‘, the firm’s code for Fermi boards, sans silicon, hacked up, and stuck together with the finest glue sticks on hand. It will even have one or two *GASP* with silicon in them! The problem? The silicon blows.
If you recall, our timelines have been accurate to the day, unlike every other site out there, and it looks like Nvidia is finally catching on the fact that its ‘official timelines’ are being openly laughed at. Now, it has finally started putting out a realistic schedule, sort of, but tries its best to downplay it. ‘Fermi’ boards for compute will be out in Q2.
The funniest part is how it pushes the gamer versions, that is, with a footnote. No really, if you look at the Nvidia link above, near the bottom, by the disclaimer, it says, “Editors’ note: As previously announced, the first Fermi-based consumer (GeForce®) products are expected to be available first quarter 2010.” Told ya.
Luckily, gaming is a priority at Nvidia, just as long as gaming doesn’t mean that there is an ATI chip anywhere in the system, then it locks you out for your own good. If you are waiting in the basement, stroking your 9500GT and waiting for a GT300 based card on January 1 of 2010, don’t. The cards have no chance of a real release in early Q1, if they make Q1 at all. Take Nvidia’s corporate honesty into account before you believe even these late dates.
Nvidia can’t yield a wet napkin with 40nm engravings on it right now. Tomorrow’s GT240 is almost a year late, underpowered, underperforming, and too expensive to be taken seriously. At $99, it gets eaten alive by ATI’s last generation, and thoroughly abused by the current one. If Nvidia is four months late with an ~130mm^2 chip on TSMC 40nm, does it really expect us to believe it can make a 530++mm^2 chip in anything approaching volume? Granted, it will make more than seven, but not enough to make a real product launch. Expect more paper and pomp than actual shipping product for this ‘hard launch’.
The worst part is the clocks. No, really. The numbers we have been hearing for GT300/Fermi for the past few months were 768 GigaFLOPS for double-precision floating-point and 1.5 TeraFLOPS in single-precision floating-point, compared to the 544 GigaFLOPS double-precision floating-point and 2.7 TeraFLOPS in single-precision floating-point for the ATI HD5870 or Cypress chip. That might not seem like a problem until you look at what Nvidia said in its presentation, that is, 520 to 630 GigaFLOPS in double-precision floating-point. A quick trip to the calculator says 630/768 = 0.82, or nearly a 20% clock miss.
Even the poster child for unprofitable chips, the G200, only missed its clock targets by about 10%, and was only seven months late. Fermi is looking like it is about as late, but 20% down on speed. Talk about starting out on the wrong foot. Even a puppy can take a wobbly step or two soon after it’s born, but Nvidia doesn’t seem to be able to do even that.S|A