The official GT300/Fermi Thread

Printable View

Show 100 post(s) from this thread on one page

10-03-2009, 11:44 AM
Chumbucket843

Quote:

Originally Posted by Calmatory

Source?

Unless it's based on the fact that when the chip size grows, less chips fit on the wafer, AND when the chip size grows, there is higher chance to have a defect on the chip.

So actually that makes some sense. Defect rate grows linearly, while chips per wafer decrease linearly. As these two add up, yields decrease expotentially. No?

wouldnt that be f(x)=2x? their is a lot more to it than that obviously. here is what dr who has to say about designing gpu's:

Quote:

Originally Posted by Drwho?

With all due respect, I don't know if you ever tried to work on a puzzle that have more than 1 billion parts ... this is what GPU are ... so, it is always easy from your screen point of view to write those kind of statements, never the less, I am sure you never even completed a Lego with 100 000 parts ;-)
10-03-2009, 11:45 AM
003

Quote:

Originally Posted by Farinorco

At least talking about supporting technologies and so (the new C/C++ compiled code support, for example). Regarding computing power, I don't have it so clear.

The only example that I have been able to find comparing GPGPU performance in equality of conditions (that includes both running the same code, so it has to be a DirectCompute or OpenCL piece), the HD5870 is pulverizing the GTX285.

And the funniest part is that said example, is an NVIDIA demo of DirectCompute (run on both cards in AnandTech). :D

To be exact, this one:

http://images.anandtech.com/graphs/r...2344/20136.png

But yes, I'm of the opinion that NVIDIA is one generation ahead when talking about GPGPUs technologies. We will see in terms of GPGPU performance when we have something else than this little demo to compare. I think it's a too specific program to make any conclusion from this.

1. I love how they leave out RV770, the real competition to GT200, and

2. That is single precision. GT300 will decimate RV870 in double precision.
10-03-2009, 11:49 AM
LordEC911

Quote:

Originally Posted by Helloworld_98

AMD must be making a huge profit on the 5870's then, 5850's too.

With a kit cost of ~$70-80 yeah they are.
Compared to Nvidia's G200 w/ a kit cost ~$120-130 and GF100 which should be upwards of $150, I think I guesstimated just shy of $200.

Quote:

Originally Posted by 003

1. I love how they leave out RV770, the real competition to GT200, and

2. That is single precision. GT300 will decimate RV870 in double precision.

As I have said before, DP is totally dependant on clockspeeds, if they hit their clock targets, they should have ~40% advantage over a 5870, if they have G200 like clocks, that can drop to a ~20% advantage.

Then what if Cypress has more units on it for the 5890 plus a higher clockspeed?
That could make it even closer.
10-03-2009, 11:50 AM
Helloworld_98

Quote:

Originally Posted by 003

1. I love how they leave out RV770, the real competition to GT200, and

2. That is single precision. GT300 will decimate RV870 in double precision.

they didn't use the RV770 because it doesn't have directcompute 5.0 which directly compares to CUDA, so it would be kinda unfair since RV770 doesn't have the same feature set.
10-03-2009, 11:59 AM
saaya

Quote:

Originally Posted by Helloworld_98

AMD must be making a huge profit on the 5870's then, 5850's too.

well, you have to think of costs for gpu packaging, gpu testing, pcb, components, memory, heatsink, testing again...

overall i think a 5870 costs amd around 200$ to make and a 5850 around 180$.
this is normal though, think of how much gt200 cards used to cost, and while yields improved and there are cheaper pcbs, that doesnt change the price THAT much... 260s are almost 150$ retail now, they used to be over 300$ iirc...
10-03-2009, 12:05 PM
saaya

Quote:

Originally Posted by LordEC911

Then what if Cypress has more units on it for the 5890 plus a higher clockspeed?
That could make it even closer.

mhhhh that might be possible actually... if you compare processing blocks of juniper with processing blocks of cypress and then compare die sizes, it doesnt match up and indicates theres actually more in cypress... but maybe they just bloated the design to improve yields or clocks et?

it could also explain the initial 2000sp rumor though...
and it wouldnt be the first time that ati launches a gpu with certain specs and a few months later yells SURPRISE!!! so far the gpu has had xx% of its logic disabled, yields are good enough to enable all of it now! :D

but idk... ill have to look at the die sizes again and check...
10-03-2009, 12:05 PM
DilTech

After much deliberation, I've decided to just keep my thoughts to myself when it comes to this thread. I'll read updated news, and post news that hasn't yet been added, but beyond that I'm not going to bother with arguing with people any more for now.
10-03-2009, 12:12 PM
Chumbucket843

Quote:

Originally Posted by Farinorco

At least talking about supporting technologies and so (the new C/C++ compiled code support, for example). Regarding computing power, I don't have it so clear.

they win in programmability which is half the battle. usually the speed of a gpgpu app is compared to a cpu. here is an interesting article about implementing molecular dynamics on gpu's.http://www3.interscience.wiley.com/c...77402/PDFSTART it ends up being the exact opposite of the directcompute app with regards to speed.
10-03-2009, 12:13 PM
Tim

Quote:

Originally Posted by DilTech

After much deliberation, I've decided to just keep my thoughts to myself when it comes to this thread. I'll read updated news, and post news that hasn't yet been added, but beyond that I'm not going to bother with arguing with people any more for now.

Same here, people post such rubbish, speculation which doesn't make the slightest of sense, fueled by a lot of hatred against nVidia. If this goes on for 3 more months, I'd rather take a complete break and check back in 3 months time. Check Fudzilla and some headlines here and there.
10-03-2009, 12:15 PM
LordEC911

Quote:

Originally Posted by saaya

well, you have to think of costs for gpu packaging, gpu testing, pcb, components, memory, heatsink, testing again...

overall i think a 5870 costs amd around 200$ to make and a 5850 around 180$.
this is normal though, think of how much gt200 cards used to cost, and while yields improved and there are cheaper pcbs, that doesnt change the price THAT much... 260s are almost 150$ retail now, they used to be over 300$ iirc...

Hmmm... I don't think those numbers are even close.

Edit- Thanks to the people that are deciding not to post anymore, means there are less posts to read through.
Amusingly enough, you posted to say your not posting...
10-03-2009, 12:22 PM
RealTelstar

Quote:

Originally Posted by LOE

It is VERY strange to see all those people who fiercely DEFEND a greedy, multi billion dollar corporation that IS OBVIOUSLY PLAYING DIRTY TRICKS not only to it's competitor, but to YOU - CUSTOMERS as well.

Think, god damn it!

Is it fanboyism? Or stupidity? Or blind patriotism or something? Damn...

LOL
10-03-2009, 01:10 PM
003

Quote:

Originally Posted by DilTech

After much deliberation, I've decided to just keep my thoughts to myself when it comes to this thread. I'll read updated news, and post news that hasn't yet been added, but beyond that I'm not going to bother with arguing with people any more for now.

I know exactly how you feel :(
10-03-2009, 01:14 PM
ubuntu83

Quote:

Originally Posted by 003

I know exactly how you feel :(

Well seeing is believing. Currently nVIDIA doesn't even have a half decent sample to show to the public.
10-03-2009, 01:18 PM
Calmatory

Quote:

Originally Posted by 003

I know exactly how you feel :(

Funny to note that both are Pro-Nvidia persons. No wonder you're feeling bad right now though.
10-03-2009, 01:40 PM
Mad1723

Quote:

Originally Posted by ubuntu83

Well seeing is believing. Currently nVIDIA doesn't even have a half decent sample to show to the public.

Well, it's a half sample, they just cut it in half :D
10-03-2009, 03:08 PM
Slovnaft

Quote:

Originally Posted by Calmatory

Funny to note that both are Pro-Nvidia persons. No wonder you're feeling bad right now though.

...sighh:shakes:
10-03-2009, 03:23 PM
annihilat0r

Quote:

Originally Posted by Calmatory

Funny to note that both are Pro-Nvidia persons. No wonder you're feeling bad right now though.

lulz

if they're "pro-nvidia", then you come across directly as an ATI fanboy; because you bash Nvidia and praise ATI way more than DilTech defends Nvidia and bashes ATI.

It's funny how you always try to act the super knowledgeable 3D engineer who knows everything and corrects every tiny mistake someone makes on XS; but then jump on the bandwagon to bash Nvidia for a product that is yet to be spec'ced, priced and released.

On a more general scale;

Currently on these forums, relentless Nvidia flaming is way more common and severe than actually defending Nvidia. What's more interesting is; those few who try to tell the incredible Nvidia bashers that they're crossing the line (of sane Nvidia flaming, which is true to a level because Nvidia indeed isn't in very good shape right now) are the ones to get called fanboys while the amount of fanboyism they're committing is orders of magnitude lower than the other group.

Disclaimer: It's been two years since I last used a Nvidia product, and in a couple of weeks I'm getting a 5870.

/rant
10-03-2009, 03:29 PM
Chumbucket843

Quote:

Originally Posted by Calmatory

Funny to note that both are Pro-Nvidia persons. No wonder you're feeling bad right now though.

this right here is why diltech even posted that. if i said i thought physx is cool why dont you just leave it at that. how is being pro-nvidia a bad thing? why is it ok to be pro-ATi?
10-03-2009, 04:11 PM
Calmatory

Quote:

Originally Posted by annihilat0r

It's funny how you always try to act the super knowledgeable 3D engineer who knows everything and corrects every tiny mistake someone makes on XS; but then jump on the bandwagon to bash Nvidia for a product that is yet to be spec'ced, priced and released.

I AM super knowledgeable 3D engineer. :yepp: Yeah, we know nothing, but that Nvidia is in deep trouble, there's no denying on that.

Just got to love how easily people get provoked here, over companies and their success, and telling how company X just plain sucks and people are fanboys. They see me trollin', they hatin'. :yepp:
10-03-2009, 04:18 PM
Vienna

Quote:

Originally Posted by Chumbucket843

this right here is why diltech even posted that. if i said i thought physx is cool why dont you just leave it at that. how is being pro-nvidia a bad thing? why is it ok to be pro-ATi?

Because they're the underdog, and regardless of if nVidias cards end up being better or not, everyone loves an underdog.

ATi is just currently the crowd favourite right now is all.
10-03-2009, 04:24 PM
Chumbucket843

Quote:

Originally Posted by Vienna

Because they're the underdog, and regardless of if nVidias cards end up being better or not, everyone loves an underdog.

ATi is just currently the crowd favourite right now is all.

how are they the underdog? they are gaining market share and they have the fastest card. by the way market share is about even between them to begin with. intel has about 50%. that still is no reason to flame nvidia. every thread that has to do something with nvidia marketing/physx/sales every ATi fanboy has to get their bashing in and the threads end up being 10+ pages long.
10-03-2009, 04:24 PM
annihilat0r

Quote:

Originally Posted by Calmatory

I AM super knowledgeable 3D engineer.

riiight

Quote:

Originally Posted by Calmatory

Just got to love how easily people get provoked here, over companies and their success, and telling how company X just plain sucks and people are fanboys. They see me trollin', they hatin'. :yepp:

Being provoked is one thing and telling someone that they're going overboard with their criticism another.

Bash Nvidia for a product that's yet to be released, priced, even spec'ced that's fine; but tell someone that that's actually a stupid thing to do, that's being a fanboy and crying to help Nvidia. Right. That's what we do. Uh huh.
10-03-2009, 04:35 PM
randomizer

Quote:

Originally Posted by LordEC911

Not going to happen with 10% yields. They didn't even have enough silicon for their PR event...

Anything is fine to "get your foot in the door" and say you've "launched" :p:
10-03-2009, 04:40 PM
jaredpace

Quote:

Originally Posted by Mad1723

Well, it's a half sample, they just cut it in half :D

Literally? :D
10-03-2009, 05:37 PM
Think

So now that the Fermi card is a fake...does this mean that Jen-Hsun Huang will perform the Seppuku? :):):)

Show 100 post(s) from this thread on one page

All times are GMT -8. The time now is 03:04 AM.

XtremeSystems