WHAT DO YOU do when there is a major international trade show, and you can’t even supply all your partners with upcoming parts? You threaten them, which is exactly what Nvidia did for its upcoming GTX480 at CeBIT.
Just before the show started, Nvidia reps ran around and made their partners sign a ham-handed NDA list of what is acceptable and not acceptable to talk about at CeBIT. Nvidia is desperately afraid that numbers like this will leak out. Even though it is too late, Nvidia doesn’t want it confirmed, so people must be shut up.
Last week, several sites contacted SemiAccurate saying that Nvidia called them and threatened to cut them off if they linked to the above story. This time, they were smart enough not to put anything in writing, but we hear there is at least one recording floating. It should be hilarious it ever becomes public.
At CeBIT, they were not nearly as tactful. Needless to say, this pissed off their partners in the extreme, to the point where two of them found the author just to show him the NDA. When you are backed into a corner and need friends, that is not the time to start threatening.
The NDA looks like this, with text below the picture.
Nvidia Rules & Regulations
Summary of Permitted Activities:
*Public demonstration of GF100 in 3D Vision using NVIDIA supplied PC and content.
*Public display of color box templates and disclosure of features and brand name information included in the NVIDIA retail box template
Not Permitted Activities:
-Public or NDA demonstration of GF100 using any other system or content than supplied by NVIDIA
-Public or NDA disclosure of performance numbers or benchmarks
-Public or NDA disclosure of clocks or any other product specifications outside of those included in the NVIDIA retail box template
-Public disclosure of pricing or availibility
-Public or NDA press briefings of GF100. Please refer all press inquiries on GF100 to NVIDIA PR.
-Public display of GF100 mechanical sample
* * *
There you have it. While it may look pretty tame, it was accompanied by some threatening language in at least two of the cases. Well done, ATI couldn’t have done more damage to the Nvidia name if they’d tried. You have to wonder about a memo that won’t use a publicly disclosed product name.
Most of the questions were answered a bit ago, but lets recap. The benchmarks are on par with the ATI HD5850 for the GTX470 and about 5 percent above the HD5870 for the GTX480, but both Nvidia cards run much hotter. Clocks are 625/1250MHz, way short of the intended 750/1500, and wattage is through the roof.
Two box samples seen by SemiAccurate had no clocks or specs on them, quite telling for a product that will paper launch later this month. There must be quite a bit of panic in Santa Clara right now. Pricing wasn’t set according to four partners talked with by SemiAccurate, and availability for the high-end part, at least according to what Nvidia has promised to deliver, is basically a few hundred units per vendor. Worldwide. Ever.
We would go to Nvidia PR for more questions, but they have stated that they see no reason to talk to SemiAccurate. Finally, the mechanical sample bit is just piling on the pain. If you read into this that each of the big name vendors don’t have enough ‘real ‘puppies’ to do their own testing, that would be the case. Months later, yields still struggle to break into the double digits for both 470 and 480 combined, and there just aren’t enough to put out, much less sell.
If you are thinking that CeBIT would have a triumphant launch of the overwhelmingly powerful GTX480, ask yourself why Nvidia has to threaten vendors, post 24/7 guards at each GTX480 demo, and not even give vendors the most basic information about what cards are being shown off in its booth. Well done Nvidia!S|A
Latest posts by Charlie Demerjian (see all)
- You can now buy bare Snapdragon SoCs with 410E and 600E - Sep 28, 2016
- Spin Transfer Technologies talks about their ST-MRAMs - Sep 27, 2016
- ARM adds CMN-600 interconnect and DMC-620 memory controllers - Sep 27, 2016
- Globalfoundries 7nm process isn’t even close to the name - Sep 26, 2016
- ARM upgrades realtime offerings to v8-R and adds Cortex-R52 - Sep 21, 2016