Nvidia Fakes Fermi Boards At GPU Tech Conference 212
fragMasterFlash writes with this excerpt from SemiAccurate:
'In a really pathetic display, Nvidia actually faked the introduction of its latest video card, because it simply doesn't have boards to show. Why? Because it didn't get enough parts to properly bring them up, much less make demo boards. ... Notice that the three screws that hold the end plate on are, well, generic wood screws. Large flat -head Phillips screws. Home Depot-grade screws that don't even sit flush. If a card is real, you hold it on with the bolts on either side of the DVI connector. Go look at any GPU you have; do you see wood screws that don't mount flush or DVI flanking bolts? ... If you look at the back of the fake Fermi, [from this PC Watch picture], you can see that the expected DVI connector wires are not there, just solder-filled holes. No stubs, no tool marks from where they would be cut out. Basically, the DVI port isn't connected to anything with solder, so they had to use screws on the plate."
Sobering (Score:2, Interesting)
Re:Lies! (Score:2, Interesting)
So for now I'm using windows XP... Bugger.
It's my brick in a box... (Score:3, Interesting)
Anyone in the embedded systems biz who's ever gone to a trade show probably knows the "brick in the box" technique.
1. You fab a slick looking enclosure for your "new product".
2. You put a brick in the box.
3. You show the box with wires coming out of it, and a PC behind the curtain displayinging the actual app.
That way, you have something to show/promise/sell YEARS before an actual product is ready, and can blame the engineers for being slow to finish and test that "last 10%".
Re:Who cares... (Score:4, Interesting)
The original CD player comes to mind. They demoed it as a small elegant device on the desk, hardly bigger than the actual CD. Under the table, hidden by the tablecloth, were the hulking electronics. But they knew that miniaturisation of the electronics would be just a matter of time and they wanted to show what the system could be.
Re:Who cares... (Score:3, Interesting)
No there was working silicon at the tech show, it was encoding the HD stream live.
Re:Ya well (Score:3, Interesting)
Just because it wasn't NDA'd doesn't mean he was supposed to reveal it. There is a little thing called "honor" and some people in the world still have it and assume others do as well. For example some time ago I was e-mailing back and forth with a guy from SVSound. He decided to let me know about a new upcoming product that wasn't public information yet (their surround speakers, which were announced on their news page a month ago). He asked me to please not go posting it on forums at that time, until they announced it on their site.
Now I wasn't bound by that, I signed no NDA, it was just some e-mails back and forth. However I did abide by it. Why? Because it is the right thing to do. He trusted me and asked me to do something, so I did.I don't know their reasons for wanting to hold on to the fact till the end of August, but they did. I respected that.
Well, similar situations can happen with journalists and companies. nVidia says "Sure here's some info on a new product, but sit on it till next week ok?" Journalist (rather loose term in this case) goes and immediately spills it to get a "scoop." nVidia say "Ok, fine, you are now persona non grata. No more info for you and you aren't welcome at our press events."
Re:Who cares... (Score:3, Interesting)
It wasn't described as a mock up, but as a real working Fermi board.
NVIDIA are quite a way behind in the next generation race (time-wise, not tech-wise), and they had to try and make it look like they were a month or two away from having product availability. This fakery just makes the late Q1 2010 rumours sound more likely...
You're right about NVidia claiming it was real
FTA: Note 1: Nvidia PR was asked to comment on the faked cards earlier this evening. Their reply was, "I'm glad you're asking us before you write. That statement is false. The product that we displayed was an actual Fermi board. The demo ran on Fermi silicon." We do not believe all of that statement.
I'm willing to give NVidia the benefit of the doubt. Sometimes hardware engineering samples are all hacked together like you wouldn't believe to get the first versions working.
The whole screw / DVI / blocked vent thing could be easily explained if they had to put a DVI connector off the card (Fermi has 3 output's) and didn't have a matching backplate for it -- the screws could be holding the "floating" DVI connector. The dual power connectors also makes sense since one is 8-pin and one is 6 pin -- they may be using the same card for more than one chip in the family and depending on power requirements may use a different power connector. He claims the power connector is fake because the mounting solder points are 90 degrees from the connector itself -- the part costs a couple pennies more but you can buy power connector sockets with rotated mountings.
A complete dissection (Score:2, Interesting)
I don't see anything particularly wrong with that card. Let's go through it in order:
1) The chip itself: If you're going to be standing on stage, potentially wandering aroung with a little piece of metal in your hand that represents your company's future and is literally worth millions to your competitors, are you REALLY going to show off the real thing? No. You use something that looks close just to indicate size, approximate appearance, and other basic details so the idiots in the crowd know what they're looking at.
2) The number 7: I'm not sure what exactly is implied by the author here, but going along with displaying basic information, it's sensible to alter the display chip with a hand-drawn logo, for the benefit of observers.
3) Blocked vent: From the look of the rest of the card, is seems that the cooling air needs to come from somewhere. My guess is that some comes from inside the case, and some comes through that "blocked" vent. See, a long time ago, humans discovered that when you run a fluid through something with tiny holes, big things (like dust) are kept out, giving you a nice clean fluid. Filters are good things.
4) Screws: Personally, I use those screws all the time in my computer. They're great for mounting disk drives, PCI cards, case components, and generally anywhere else you need a small machine screw. Screws don't always fit perfectly in a final product, let alone in a prototype.
5) DVI bolts: Yep. All my DVI ports have them.
6) Stacking two single-slot cards together to show the end plate doesn't need screws: Single-slot end plates are most likely held on with screws through flanges bent over the card itself. How exactly would that be easier on a double-slot card than just punching a hole and running some machine screws through, especially considering that there's no indication of what's actually behind those screws? My personal hypothesis is that the screws go into a plastic wall that divides the card,
7) Soldering of DVI port: Personally, I think it'd be easier to just solder a DVI port in than clip off all the little wires from the port so it would physically fit where it is. This entire claim is based around apparently a single photograph. Judging from that same photograph, there's also no contacts on the edge connectors, and only a smudge written on that sticker in the middle.
8) Half-covered SLI connector: SLI is an edge connector. From the same photograph, it appear's there space there to make contact with an edge and cover the contacts with a thin connector. That should work, right?
9) Power connectors: Assuming that it's absolutely impossible to use wires to connect anything over a distance of one inch, the 8-pin connector appears to just rotate its pins by 90 degrees, probably to accommodate the other stuff that appears on the board in the immediate vicinity. I'm not an expert, but according to this [allpinouts.org], that extra connector appears to provide just more power, so would it not be possible to connect it in the port, rather than on the board?
10) Glue: Since when is glue not an acceptable means of attaching parts? I personally have used glue many times, in many ways, for the purpose of holding things in place. On a board destined for display, it seems like an even more practical solution.
11) Board being cut off: This one almost seems legitimate,except for the fact that I have a few boards lying around here with traces (and a few components) right on the edge of the board. Yes, it looks a little crude. In fact, it almost looks like a prototype made for display, possibly even by just cutting off test circuitry from the board.
12) Exaggerated marketing: I'm really not concerned that the spokesman said "This is Fermi" if it isn't. The point is that it's a close approximation, and the card's actual functionality isn't an issue. If they had used a real prototype that happened to burn up during testing, it would be hailed as evidence that "the car