- Friday, 13 November 2009
- INQ Mobile
- RSS
It was only a matter of time before the INQUIRER found out - Top Intel spinner
Nvidia 55nm parts update
NVIDIA IS SET to trickle out the latest batch of 55nm parts. Expreview has some pictures and tidbits about the latest 55nm GT200/GT200b here, and some GX2 info here.
It looks like the on-again, off again GT200GX2 is on again, and it is called the GTX295. Yay. The 55nm parts, internally code named GT206, are finally trickling out like we said they would, with no speed increases, and no power gains. What should have been a simple optical shrink is turning into a totally botched job, with the 'real' 55nm parts unlikely to come out until late January at the earliest following yet another spin.
Given the lack of gains with the B2 stepping, the GX2/GTX295 still seem unmakable in volume, but such trifling concerns have never stopped Nvidia in the past. We hear they are going to launch it even though they can't make it, along with the requisite 19 parts to Newegg so they can claim it is on sale. Real volume won't happen until (if?) they can fix the power problems.
We hear that the 'launch' is likely going to happen at a shindig on the 12th of December so they can claim the win they promised before the end of the year. One has to wonder if cherry picking parts in an attempt to use tame press to snow the public is the definition of 'Whoop-ass'? I am sure they will claim a stunning victory in any case.
One way you can tell how screwed up the chip is is the use of a heat spreader and a stiffener (the metal ring around the chip). If you have a big die, you need mechanical support for it, or it can crack or break bumps. A stiffening ring is usually the cheapest and most efficient way to go, but in many cases, a heat spreader will do the same job.
The problem with a heat spreader is that it introduces two additional thermal barriers, the paste under the lid and the lid itself, to the cooling of the silicon. Each one makes cooling incrementally less efficient, not to mention material and assembly costs. You don't do this unless you have to.
If you are wondering why every modern CPU out there has one, the answer is simple, so ham-handed monkeys like most DIY people don't crack the die when they clamp the heatsink on. Think AMD K8 here. CPU makers think the cost of a spreader, and the reduction in performance it brings, is worth the protection it gives.
GPUs however come assembled. Factory robots don't break chips, so the mechanical protection is not an issue, but the costs remains. So, why did Nvidia do it on the GT200? They can't control hot spots. The lid is a heat spreader, and it helps keep chips with poor hot spot control alive and working.
When you see a heat spreader on a part that comes assembled, it is a pretty sure sign something is wrong thermally, it simply is not worth the cost and performance drop otherwise. Make no mistake, the spreader and stiffener combo on the GT200b is a bad bad sign.
Why is the GT200b such a clustered filesystem check? We heard the reason, and it took us a long time to actually believe it, they used the wrong DFM (Design For Manufacturing) tools for making the chip. DFM tools are basically a set of rules from a fab that tell you how to make things on a given process.
These rules can be specific to a single process node, say TSMC 55nm, or they can cover a bunch of them. In this case, the rules basically said what you can or can not do at 65nm in order to have a clean optical shrink to 55nm, and given the upcoming GT216, likely 40nm as well. If you follow them, going from 65nm to 55nm is as simple as flipping a switch.
Nvidia is going to be about 6 months late with flipping a switch, after three jiggles (GT200-B0, -B1 and -B2), it still isn't turning on the requested light, but given the impending 55nm 'launch', it is now at least making sparking sounds.
The real question is, with all the constraints and checks in place, how the heck did Nvidia do such a boneheaded thing? Sources told us that the answer is quite simple, arrogance. Nvidia 'knew better', and no one is going to tell them differently. It seems incredulous unless you know Nvidia, then it makes a lot of sense.
If it is indeed true, they will be chasing GT200 shrink bugs long after the supposed release of the 40nm/GT216. In fact, I doubt they will get it right without a full relayout, something that will not likely happen without severely impacting future product schedules. If you are thinking that this is a mess, you have the right idea.
The funniest part is what is happening to the derivative parts. Normally you get a high end device, and shortly after, a mid-range variant comes out that is half of the previous part, and then a low end SKU that is 1/4 of the big boy. Anyone notice that there are all of zero GT200 spinoffs on the roadmap? The mess has now officially bled over into the humor column. µ
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
I'm sticking with my 8800 GTX till they release the GTX 300 Series.

I will buy two if Lucidlogix has implemented their Hydra 100 chip by then, either on the graphics cards themselves or in an expansion pod that I can hopefully squeeze in my case.
It's always delightful to read Charlie's "news", lack of facts, complete misunderstanding of anything related to physics and other fun stuff.

While heatspreader will do it, what it says, the same thing would be applied to heatsink. If machines don't break them, then the heatsink would spread the heat to a wider area, thus making it easier to cool it.

That certainly isn't the reason with hotspots and heatspreaders.
............. Hummn another failed by Nuked-Vida(vida in hindi means gone again nuked in some sense) article.
Here goes our charlie.


However.. I have one job for you.. 
Report on what happened between the Bedsharing Agreement between Nvidia and Intel Crosslicensing.
how badly this agreement is failing too.
Which was provoked by ATI-AMD merger.


Regards
mohnkhan

Mohiuddin Khan Inamdar.




Just because the facts back up your assertions you think it is ok to go around bashing a company that has demonstrated time and time again its total disregard for its own customers in deference to its bottom line? For shame, Inq, for shame.
You know absolutely nothing about heat spreaders and what purpose they serve. I don't work for Nvidia but I have spent more than 16 years in Semiconductor Assembly Test and you are absolutely wrong. As the DI sizes shrink there is more need to spread the power/heat out because it radiates from an ever shrinking DI surface area. You ever wonder why laptop chips are always lower power yet have no heat spreader. You need to go back to school and learn more about thermodynamics buddy. You contradict yourself in two different paragraphs by saying that heat spreaders introduce a thermal barrier yet later you say that they diffuse hot spots on a DI. Which is it Monkey boy? Do they more effectively spread the heat for a hot chip or do they impede thermal conduction. Everything you said in this article contradicts itself!
The end user doesn`t give a damn about heatspreaders or metal rings. As long as products work and bring high fps and are payable he will buy them- easy as that.
Your constant nvidia bashing is a joke in itself and nobody takes it seriously anyway anymore in "the scene".

Nvidia still has the performance crown with triple GTX280 (yes, they are faster than 2x 4870X2) and the new GTX295 "GX2" card in SLI (=Quad SLI) setup will kick the crap out of Quad CF- and you will have to live with it - HAHA ;) 
I think this re spin and change to 55nm, wil just be updating to the 55nm process, fixing all this issues with failing /faulty chips."I hate Microsoft"

no offence charlie but after the first 10 lines, I loose interest in your long winded spiels about how some company is screwing ppl over blah blah , "I hate Microsoft". 
so if it seems if i am misinformed it is because the piece was too long, and quite frankly I don't have the time to read your stuff, since you don't have the time to write about seriously customer screwage,"I hate microsoft" that I have pleaded for your help on so many times
'clustered filesystem check'

That really made me laugh! Very good work, Charlie!

And, thanks for the low down on the new nVidia chips. It makes me sad though - I like CUDA a lot so I tend to buy nVidia cards. I was hoping the new rev would solve the reliability problems with the 200 chips.
I know I know Charlie...Nvidia is going out of business...LOL. Wait, that's right, I think you said they would ALREADY BE out of business by now.

Give it up man. They're kicking AMD/ATI's a$$ in the money dept. At the end of the day the only thing that matters is "are we making money?". NV's answer is a resounding YES. AMD/ATI? NOPE. For 2-3 years still NOPE?...I rest my case.
Quite possibly the reason is that GT200 is basically a beefed up G80. They already have a low-cost version of that, it's called G92.
Just a quick note: Not all modern CPUs have heatspreaders - laptop processors being the most obvious exception. This is for three reasons: the two you mentioned (idiots generally don't replace their laptop processor, and thermal performance is vital in a laptop) and also because a heatspreader adds another 5mm or so of height, which causes problems in todays thin laptops.
Starting to see a disturbing pattern in these articles. Did Nvidia refuse to send somebody free samples for testing?

As a Canadian I prefer ATI over Nvidia but own an Nvidia card right now. 

Anyways maybe this site is trying to be the Howard Stern of IT news, I know it's cool to be controversial these days but this is just funny!!! 
If CNN & Fox News had a child it would be named The Inquirer!
Nvidia couldn't be in better position, with DAMMIT going bit under. Yet, Nvidia instead chooses to remain "above" Ultimat. Or incompetent.

Gtx 350 mustn't be too far out in '9, as 295 indicates that 290 is being skipped & new plan to jump to next level is being implimented, its XX5 at end, in days of 4 digit numbers, XX50

?GTX 295, should be plow blade to faster future, yet it might be another bungled xp product. it took six weeks just to decide upon letters GTX earlier in year, So Now, perhaps most powerful thing Nvidia could do is make clear shot at NT6, Ultimate crown or merge with AMD.
STeWie Drashek
There is no need for GT200 derivative parts, as the new series did not really bring any new technology to the core. The 8xxx or 9xxx series are perfect derivatives already, it's not like a GT200 derivative would add anything.
Dear Charlie, 
How would your "story" about all the delays deal with the fact that the expreview chips are manifactured in week 33?
And tell me why Jen-Hsun Huang said that the transition to 55nm was completed in Q2, and in Q3 only 55nm chips have been manifactured? (http://seekingalpha.com/article/104608-nvidia-corporation-f3q09-qtr-end-10-26-08-earnings-call-transcript?page=-1)
It seems to me that nvidia would be in no rush to release a part that would butcher sales of the 9600/8800/9800 series which there is still piles of stock to get rid of.
They will prob let these cards be the value/midrange line-up for some time to come. Since the gtx200 series brought no arc. improvements to speak of (dx 10.1 or better) just sheer performance- what would be the point of created abother card to fill a gap thats not there? the g92 cards fill their price range nicely. Cut down the gtx200 and what do you get?
A g92..... 
what would be the point?
The point is moot cause most are buying amd despite all the recent core 216 spin... too bad the core216 scores don't help them sell 9800's....
Gotta admit that the old core gtx260's are lookin like a tasty deal at or just above $200.
Hard to argue with 2 x 4830's for just a bit more though! (beats the 280)
Please state your source or everyone will just think you are talking out of your A%%.
Why nVidia would make a half-hearted GT200? They already got the G92b parts running happily in 55nm, and those are really a GT200 cut in half. And the G94? A GT200 cut in 4, just like you wanted.
The GT200 is just a bigger G92, there are no compelling architectural enhancements to justify the spin-offs. That's why they love to rename their chips!
All ATI need to do now, is cherry pick a 4870, hand it to the engineers, and overclock and overvault it, and put the fan speed up considerably. 

If they can cherry pick a part that will match the "gtx 295" all they then do is have a competition, all around how they are actually faster, and give that one part away.

That way the press dont rip them apart for a paper launch, but they can still claim the performance crown.

Turn that back on nvidia.