"AMD - 5870 is better than fermi"

[Greebo]

As for what it was on that starred out part of my earlier post... It was cake. A nice sweaty fruitcake I made last week. I'm eating it now, so you can all be quiet

i'm glad it was cake

sorry for having a go at u earlier, it was meant it the best spirit

 

[BubbySoup]

Cake ...that's a load of **** and you know it!

 

[madindehead]

I thought it was Nvidia who have themselves stated that the Fermi has 8x the single precision power of the GTX280?

Nvidia stated is has 8x the DOUBLE precision power of GT200 series. Not single.

Either way, read something interesting in Custom PC mag today (my first issue arrived today):

Interestingly, Nvidia has chosen not to talk about the graphics portion of the chip, which has led to concerns that Nvidia is no longer focusing on gamers. Conversations with top executives confirmed that graphics is still very much at the heart of Nvidia and we'll hear more about the chip's graphic capabilities in due course.

I wish people would stop writing off Nvidia. Fermi is a departure yes. Into a new area for the company, so that if PC gaming does die, the company doesn't die with it. Do you really think they are just going to walk away from PC gaming altogether when they know it is still a good market for now? No.

Let's just sit it out, and see it happens.

 

[Duff-Man]

Cake ...that's a load of **** and you know it!

Are you implying that the cake is a lie?

 

[drunkenmaster]

Theres almost smeg all difference between a "gpu" and a "gpgpu", and all the talk suggesting otherwise is rather silly.

If you want to do tesselation in an accerated piece of hardware, or in software on non dedicated hardware, is just a choice, they weren't identical architectures before(AMD/Nvidia) and they aren't now having their own choices being made.

We don't know for sure how much less of a "gpu" it is yet anyway. It may well have everything in dx9-11 hardware accelerated, it might not, it might not and still do them very well in software, or it might be utterly crippled doing it in software and suck big time in newer games. I would suggest something in the middle, a competitive card that takes a bigger hit than the 58XX through to 3 generation old parts of ATI's when using tesselation and other things, it probably won't suck though.

The end result isn't the interesting factor or the important one though, as a company, the viability or selling millions of cards at cost to maintain sales, well, isn't very good long term. Fermi, slow or fast, doesn't look like it can realistically turn a profit and thats really what Charlie et all are talking about. Charlie hasn't bashed it saying its slow, it probably won't be, but he can't see how a 50% larger core with far more complex PCB can be sold competitively priced against AMD parts. is 5-25% more speed, worth 25-50% more in cost?

Will a 395gtx with two cores blow away a 5870x2, probably, will it be required, no, will the 5870x2 be able to be sold at a higher profit than a 380GTX? These are the important questions when thinking about the "future" of Nvidia as a gaming company, and they are valid questions that have worrying answers.

How many people on this forum would pay £400 for a Fermi, when a 5870 offers 80% of the performance and is dropped to £200-250?

Its not just is Fermi better than the 5870, its how much better at what cost.

Maybe Fermi can beat it and Nvidia screw AMD on pricing by selling it at £200 aswell, but how long can Nvidia do that?

 

[bru]

your absolutly right, the whole issue will come down to the price they push it out the door at. whether its faster by a lot or not faster at all, its that all important price that counts. in this currant ecconomic climate ati priced thier new range higher than a lot of people expected. how much wiggle room they have is anyones guess,will nvidia be able to bring fermi in at a realistic price we will have to wait and see.

 

[NickK]

Exactly - hence my concern over cashflow.

Tegra may replace the source of cash from low-mid range cards over time but during the transition they're fighting on two fronts.

AMD's strategy by bringing the 58xx to market means:
a) forces nVidia's hand to reveal their next gen earlier than they wanted (causing PR issues as they're not ready).

b) Having price-performance that undermines the Fermi, making it seem an expensive product to the graphics market when evaluated on graphics only performance.

c) AMD's drive at physics with havok and bullet, undermine the unique selling point of Fermi being a dual purpose card (graphics & physics). The move means that nVidia seem to be charging a premium for generally available functionality.

d) Laptop OEM move quicker to AMD rather than stay on the dead end nVidia product lines. Once that decision is made, usually it's a while before the next opportunity exists to switch back.. that time period will be a painful reminder if nVidia want to maintain a foot print in that market.
I get the feeling that nVidia underestimated the speed of this change and how quickly stock would diminish forcing vendors to Intel/AMD.

However this war will be one or lost with the software used by developers. Something that nVidia have a head start on but one that will compete with MS, Apple and other OpenCL tools will appear over time.

 

[TAC4U]

nVidia is certainly taking a more risky approach with this generation but that doesn't mean it won't pay off. However, ATi managed to beat them to the punch with this generation and with specs that compare very well based upon the reasonable estimates made. nVidia can't really afford to bump up specs much as that will further reduce yields, so that limits their options somewhat. I expect more focus on initiatives like PhysX, 3D Vision and CUDA, and more developers being massaged through the TWIMTBP program.

 

[bru]

b) Having price-performance that undermines the Fermi

until the performance and price of fermi are actually known that is pure guesswork.

c) AMD's drive at physics with havok and bullet, undermine the unique selling point of Fermi being a dual purpose card (graphics & physics). The move means that nVidia seem to be charging a premium for generally available functionality.

again cant really quantify that until the price is released. looking at last generation generally the 260 was priced pretty much level with the 4870 and the 275 again matched the 4890. or at least until the 5800 series was right upon us.

 

[NickK]

until the performance and price of fermi are actually known that is pure guesswork.

Yes, and the majority of business is calculated guesswork. Market data is an estimation, a business case is based on fact and numbers based on the number of units sold which is based on an estimate of what the market will buy (perhaps based on previous estimates and results).

1. We have the stated FLOPS performance from nVidia themselves
2. You're right, we don't know the real world performance as nobody has had an engineering sample to bench yet. Yes we're assuming a constant based on linear scaling of the FLOPS measurements.
3. Given the maximum price that the market bracket will bear - that is known. The product needs to fit that price.
4. Given that we know, based on umpteen generations, that the fabrication is high cost compared to average.

again cant really quantify that until the price is released. looking at last generation generally the 260 was priced pretty much level with the 4870 and the 275 again matched the 4890. or at least until the 5800 series was right upon us.

My point was about the utilisation of a unique selling point to justify the additional cost. Yes you're right that we don't know that exact cost however we do know that nVidia can't really sustain making a loss on the cards.

 

[sargatanas]

again cant really quantify that until the price is released. looking at last generation generally the 260 was priced pretty much level with the 4870 and the 275 again matched the 4890. or at least until the 5800 series was right upon us.

The GTX260 was about 40-50% more expensive originally than the 4870 and they were forced to bring it down when they saw it couldn't compete at that price range, it was dropped massively.