- Saturday, 10 January 2009
- INQ Mobile
- RSS Feed
What a world! What a world! - The Wicked Witch of the West
So, who will design AMD graphics in the future?
WHILE Nvidia's Jen Hsun Huang sometimes gloats a tad rudely over the competition's headaches, there could be one point justifying his recent analyst talk confidence - the leapfrog game between Nvidia and AMD-ATI is expected to continue as usual between the HD4870 series, then GeForce 9900 series, then their respective 40nm shrinks early next year.
It's business as usual, with the two parties doing a regular leadership switch.
What about deeper into 2009? Yes, you'll all mention Intel Larrabee - after all, that is Nvidia bosses' favourite topic these days it seems. They don't even seem to mention DAAMIT much in that contest. Why?
One reason seems to be that steady stream [Shurely 'flood'? - Ed] of departures from the GPU division of AMD. Are they saying "Daamit, I had enough of this"? Whatever the case, it won't affect the R700 series and probably the 40nm shrink either. But, whatever the roadmap in place, future GPU designs can't be done without the right people.
This could be bad news if not properly addressed. I happen to quite like the current batch of ATI GPUs - while they may not be the highest performers, they are pretty fast still for comparatively low power consumption and, of course, price.
Also, they handle desktop graphics well, and now OpenGL looks better with the new drivers. The branching-capable architecture supports computational GPU work for the past three generations, and the RV770X2 will provide very nifty dual GPU shared memory system, rather than the usual duplication, that will be even more useful in computational work then Crossfire gaming. For example, two GPUs could work in parallel on a single problem and dataset, just like a dual-CPU system does today.
Also, despite the eternal Intel-AMD rivalry, the Crossfire parallel GPU operation continues to be supported on most Intel platforms. The X38 and X48 support it today - as do mainstream P35 and P45 chipsets. With Nehalem's arriv al in five months or so, both the X58 TylersburgDT desktop and TylersburgDP dual-CPU workstation chipsets will support 2 x PCIe x16 v2 Crossfire operation right out of the box. So, dual RV770X2 cards might be seen in many early Nehalem gaming rigs - especially since there's no news of an Nvidia Nehalem chipset yet.
What could AMD do next? Let's assume the worst, and no replacement of all those lost people. If fresh blood doesn't come in, one solution, in my mind, is merging the GPU group into the CPU one, and working out a new strategy where one or more R700 like units are integrated into future high-end CPUs - if AMD stays in the high-end CPU business long enough, that is.
A CPU with four X86 and two RV770 cores on a single die in 32nm process might not be that unviable, if the memory bandwidth issues are taken care of (quad channel DDR4 or something). With minor feature updates along the way, that would make the R700 series work useful for at least two generations of high-end Fusion GPUs. That could also be an interesting answer to Larrabee.
Even without it, as we can see, the Intel platform advantage, coupled with 40 nm shrink, can help the AMD GPU group survive the next nine to twelve months just fine. But beyond that, a major recruitment drive of design teams and associated positions is vital.
Put simply, too much brain drain lowers the collective IQ, and we don't want Nvidia to aim for near 100 per cent discrete GPU market share in late 2009, with only Larrabee standing in its way.
Now, was that another reason for Nvidia not exactly liking it?
Share this:
Business Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
You have a much too short-term view of processor development. It takes four years to develop a CPU, somewhat less for a GPU, but not much less (if you don't count shrinks and minor updates that happen every year). A brain drain now will not affect R700, probably not R800, but somewhere after that it starts to show (and according to an earlier article by theinq, even the specification of R900 was closed by last fall). And you can't help the situation by hiring more engineers a couple of years down the line. The work of those engineers would start showing up only after several years. There is too much criticism of CPU and GPU companies for not reacting fast enough to the performance advantage of their competitors. You have to understand the time scales, and even the company that has the advantage keeps moving forward.
Is it practical, in a multicore CPU, for some of the cores to be CPUs and others to be GPUs? A quad core with two of each type would be ideal for most mid-range computers. AMD, with ATI in-house, seems perfectly placed to do this.
"Are they saying "Daamit, I had enough of this"" --- Charlie Demerjian : THE DAAMIT SALESMAN "I happen to quite like the current batch of ATI GPUs - while they may not be the highest performers, they are pretty fast still for comparatively low power consumption and, of course, price." --- Does it mean that you like a product that offers lesser than what you have paid for?
and it probably means the end of this company, longer term. Sure, AMD can integrate their latest high end graphics part, and it will be entry level performance by the time it comes out. Then what? AMD's purchase of ATI, and the consequential financial issues that resulted, timed on the back of a poor CPU execution has made a formidable competitor weak. Hector's brilliant plan to integrate graphics has demotivated the entire graphics group at the former ATI and we're seeing the results today in this exodus. Can't really blame them. And you have to wonder how many of them, with their specialized training and not a lot of employment options, are ending up on nVidia's door step. Asset light, split the company, blah blah blah. AMD is done -- by the incredible mis-management of Hector's own hand. They tried to do too many things off the back of K7's success and have overextended themselves across multiple areas. AMD will be around to help Intel drive the x86 into full commodity (cheap) mode. Then what? buh-bye
Now listening to my story of a guy name nVidia 3 Teraflops but WE all know it's really nothing more then Tri Sli,.. A poor moderately Dow low,.. ...And ATi barely can keep it's engineers team for being canned,.. Then one day DammiT was looking at tha next 2-3 gens CPGPU or GPGPU,.. And up through the years came a moniker' SRM6 (ATi GPGPU SpeedRacerMark6). 3Gens GPGPu that is, black gold, A Green Tea Partay!,... Well the first thing you know ol' AMD a profiteers, Jen Hsun Huang's said ATi move away from there,.. Jen Said 3 Tri Sli speedup is the place you ought to be!,.. So ATi AMD (DaMMiT) loaded up the engineers and moved to A Flexible Kernel for Adaptive tha Next Gens GP+GPU,.. Without any preprocessing or additional topology data structure, that is,.. Refinement engine pools, Quantum starrs,.. The ATi & AMD DaMMiTies!,.. DaMMiT that is. Set a spell. Take your shoes off. Y'all Bishes Better Buy are Newer Shyt Now, y'hear?!?
If you are going to spew hate, at least try to be accurate. Your hate is blinding you. Please, try reading the byline. Charlie did not write this article. It was written by Nebojsa. Best regards, Daniel R.
Long Gone are Times of GL Introduction, first GL IDE, improvement of which there just couldn't be enough. So in third Milleniums' Birth: Open GL, which leads to Today. Fusion was Nvidia Product that seems to have migrated to Ati. First getting Vista ultimate Store Item might help Define Next GPU Advances.Or hide Underneath Charles Desk & Sell Pirate XP Discs? Obviously Open GL, with its improvements throughout mainboard have Won & ATI has accepted that. It Took that change Nearly Decade for engineering to take place.Meaning stepping up Now is from Best position. Somehow in ferment of 100/150/200/300 & integrated developement HOT TOP End Cars with Pleased as Punch Ultimate Smile will appear. Phophets get away with it By Taking Very Long NAPS Between Reincorporations. drashek
Never see a chip company bounce back after the great fall. The biggest problem is losing talent and that causes product delay and then it loses more talent. It’s a bad spiral no technology company can get out. IP is nothing once the talents who design the IP are gone and they will leave in a heart beat, either to join a better company or start their own.
<strong>It suppose to be Intel</strong> <br></br> AMD eggs on NVIDIA's war
Unfortunately AMD's CPU problems are stangling the finances for R&D for ATi. As an engineer would you prefer to work on nV's old hardware with old ideas, or ATi's new tech (new DX, new memory designs, new media processing, etc.) or nV's stuck in the past? AMD's problem isn't the tech or their products (which unlike the tools thing, may not be the top performer, but are still solid values and sell well [and don't have the yield problems ofthe G92]), AMD's biggest problem is that they will never retain engineers when they are tightening the belt. To save themselves AMD need to spend money on R&D and stop wasting it on advertising in NASCAR (who thought that was a good tech play?). Advertising underperforming products doesn't help, designing efficient mid-range products would generate exponentailly more money than winning the eWang competition for the boutique overpriced cards. It's nice to be king of the hill, but it the low and mid range that make all the money. AMD keeps hammering away at that market share and then they should have the money for more R&D. The past of using the high end to test features to be brought to the low end may be over, and if I were looking to jump ship from AMD/ATi I certainly wouldn't be tying my hopes to the sinking nV ship while they try to convince people that GPUs are the way of the future while running their presentation on X86/A64 rigs. If you're a quality enginee you're likely looking to intel, not nV for your next job. Only those stuck in the raster-limited past would feel the need to move to nV as the keeper of the old guard methods. Those looking to be the very cutting edge would look to open their options not close them.
Even While Cutting old Chord Out, Tra La Style, AMD most Certainly Must be Snooping for Anyone with PHYSX Experience & Recent Kill Button installed. It'd SAVE ton O' money To Grab ONE, So Keep Your Eyes Open, Don't Kill it Yet, All that NEW Finalized NT5 Stuff for Open Programable memory from Open GL (Created name: AOpen as complete lineof Desktops that Actually WON Ultimate Sweepstakes, as it turned out) to Todays GL.Getting Nvidias Physics cann't be wrong. Signs: People Whom Seem To Know Something, NOW With mikes New HUB Base for HQ, Keep Em there & Cough Up Info, Bub. AMD Is hiring, its just question of Whom You Maybee. Stewie drashek
Is this a question of sport? Don't look at me_ I'm going to be on the sick... "These eyes The hurtin's on me And I will never be free, no" Doctor Spinola, my eyes "Tell me what is wrong Was I unwise to leave them open for so long? People go just where they will ... That its later than it seems Doctor, my eyes Tell me what you see I hear their cries" And some of them sound so mean The kids are alright: Ruby, Hector, Jen and Otellini. But the rest of this lot is waking the dead. GB? Is he a Bee Gees fan? I'm sure Alan Sugar and the Beeb can find an apprentice and the career opportunity of a lifetime working for Hector AMD gfx. Pounds to Pesos, you bet your stake.
"If you're a quality enginee you're likely looking to intel, not nV for your next job" It's actually the other way around. Working at Intel is a very bad idea unless your mediocre. Here is why: good engineers get stock options and other benefits tied to the stock, working for a company like Intel you get negligible benefits out the stock. At a company like NVIDIA with a smaller market cap and a much more volatile stock it's not hard for a normal engineer to get > 200000 a year if timed correctly. Working at a company like Intel, that would never happen. On top of that, Intel is such a large beast, much more political than most smaller companies and down right depressing to work for.
Look, we all know mr.salesman gets hurt whenever we say something to the all-so-divine Hector.