Laptop graphics charts.

Why hasn't this been stickied? Its the most helpful chart on the board.

 

There are a lot of stickies already up there; any more and it would be too crowded. There is a link to this thread from the Mobile Graphics Card Chart, that's all I can do for now.

 

hmmm why not remove a few stickies then and add this to another of Your stikies instead?

Bumping seems worthless and this thread really answers a lot of those q's people have

 

fyi ive got a Dell E1505 with the x1300
stock speeds on it are
Core:400mhz
Mem:350mhz

 

i got a Go6600
Core: 375 / Mem: 300 (DDR to 600)

 

Heh, cheers guys, eyuras and deedeeman I appreciate your info but the clocks are always going to be approximate since it varies from model to model. It would be impossible for me to list everyone I came across. If I am having difficulty finding info then I will point it out (its often enough which is sad really). Nice to see that both yours are higher than those listed (which is the way I would like the list to be).

I will say if your clocks are LOWER than those on the chart and I would be interested.

 

Ultra High end
--------
7900GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTXgo = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTXgo = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
X1800 = 16 pipes, 8 vertex units, 256 bit mem (450/500) SM3
7900GSgo = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3
7800go = 16 pipes, 6 vertex units, 256 bit mem (350/550) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 12 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Added all the sections where there is a decent performance gap.

 

Quick correction -- the Go 7600 has 8 pipelines, not 12.

Refs:
http://www.laptoplogic.com/news/detail.php?id=541
http://www.mikeshardware.co.uk/RoadmapQ106.htm

Amazingly, it manages to keep up with the x1600, even with 4 pipelines tied behind its back!

 

Ah I knew I had forgotten something, Nvidias documentation provides incorrect fillrate information.

It's not too surprising, have you seen how much the 7600gt whips the x1600xt?

Ultra High end
--------
7900GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTXgo = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTXgo = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
X1800 = 16 pipes, 8 vertex units, 256 bit mem (450/500) SM3
7900GSgo = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3
7800go = 16 pipes, 6 vertex units, 256 bit mem (350/550) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 8 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Added all the sections where there is a decent performance gap.

 

Ultra High end
--------
7900GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTXgo = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTXgo = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
X1800 = 16 pipes, 8 vertex units, 256 bit mem (450/500) SM3
7900GSgo = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3
7800go = 16 pipes, 6 vertex units, 256 bit mem (350/550) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 8 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 pipes, 1 vertex unit, system memory only (475/N/A)** SM3
6100 = 2 pipes, 1 vertex unit, system memory only (425/N/A) ** SM3
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Added all the sections where there is a decent performance gap.

 

um dont forget the new nvidia go7900 with 512 mem

 

Well the core will be the same, I do not have the time atm to find the clocks (which could be the only helpful difference) but 512mb for a single card is not THAT helpful yet, maybe next generation.

 

As far as I can tell there is NO difference.

 

Why isn't the x200m included on the chart?

 

It is included, its under Ultra Low End listed as xpress200

 

Looking too high?

 

Hi! Nice list!

Anyways, you probably already know this, but the 7800 go in the Dell's are clocked at 250/658 (or would that be 250/329 on your list?)... It may affect your list, because as far as I know, no other manufacturer offers the 7800 go.

Please correct me if I'm wrong

 

Your right they have clocked it right down and they are the only ones who offer it.

Ultra High end
--------
7900GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTXgo SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTXgo = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTXgo = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
X1800 = 16 pipes, 8 vertex units, 256 bit mem (450/500) SM3
7900GSgo = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
7800go = 16 pipes, 6 vertex units, 256 bit mem (250/329) SM3
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 8 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 pipes, 1 vertex unit, system memory only (475/N/A)** SM3
6100 = 2 pipes, 1 vertex unit, system memory only (425/N/A) ** SM3
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Added all the sections where there is a decent performance gap.

 

Ultra High end
--------
7900GTX SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTX SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTX = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTX = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
X1800 = 16 pipes, 8 vertex units, 256 bit mem (450/500) SM3
7900GS = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
7800 = 16 pipes, 6 vertex units, 256 bit mem (250/329) SM3
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 8 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 pipes, 1 vertex unit, system memory only (475/N/A)** SM3
6100 = 2 pipes, 1 vertex unit, system memory only (425/N/A) ** SM3
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Ive noticed some people using this list elsewhere, please link to this thread, mostly to ensure people can stay up to date.

Removed "go" from all of the cards to clean it up.

 

..X1800 has only 12 pipes.

 

It has sixteen pipes.
http://www.ati.com/products/MobilityRadeonx1800/index.html

 

No, X1800XT has 16 pipes and X1800 has 12 pipes as I said. They are different cards. Scroll down that page...

 

--------------------------------------------------------------------------------

Ultra High end
--------
7900GTX SLI = 24 pipes, 8 vertex units, 256 bit mem (500/600) *2 SM3
7800GTX SLI = 24 pipes, 8 vertex units, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTX = 24 pipes, 8 vertex units, 256 bit mem (500/600) SM3
X1800 XT = 16 pipes, 8 vertex units, 256 bit mem (550/650) SM3
7800GTX = 24 pipes, 8 vertex units, 256 bit mem (400/550) SM3
7900GS = 20 pipes, 7 vertex units, 256 bit mem (375/500) SM3
X1800 = 12 pipes, 8 vertex units, 256 bit mem (450/500) SM3

Mid-High end
--------
x800xt = 16 pipes, 6 vertex units, 256 bit mem (480/550)
6800U = 12 pipes, 5 vertex units, 256 bit mem (450/550) SM3
x800pro = 12 pipes, 6 vertex units, 256 bit mem (400/400)
7800 = 16 pipes, 6 vertex units, 256 bit mem (250/329) SM3
6800 = 12 pipes, 5 vertex units, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 pipes, 5 vertex units, 128 bit mem (470/470) SM3
7600 = 8 pipes, 5 vertex units, 128 bit mem (~450/500) SM3
x700 = 8 pipes, 6 vertex units, 128 bit mem (350/300)
6600 = 8 pipes, 3 vertex units, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 pipes, 2 vertex units, 128 bit mem (432/400) SM3
x1300 = 4 pipes, 2 vertex units, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 pipes, 3 vertex units, 64 bit mem (450/450) SM3
x600 = 4 pipes, 2 vertex units, 128 bit mem (400/250)
x600se = 4 pipes, 2 vertex units, 128bit mem (358/277)
9700 = 4 pipes, 2 vertex units, 128 bit mem (450/213)
9600 = 4 pipes, 2 vertex units, 128 bit mem (333/? (240 turbo))
6400 = 4 pipes, 2 vertex units, 64 bit mem (400/350) SM3
7300 = 4 pipes, 3 vertex units, 64 bit mem (350/350) SM3
x300 = 4 pipes, 2 vertex units, 64 bit mem (300/230)
6200 = 4 pipes, 2 vertex units, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 pipes, 1 vertex unit, system memory only (475/N/A)** SM3
6100 = 2 pipes, 1 vertex unit, system memory only (425/N/A) ** SM3
xpress200 = 2 pipes, 2? vertex units, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 pipes, (Array) vertex units, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Ive noticed some people using this list elsewhere, please link to this thread, mostly to ensure people can stay up to date.
Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Ive noticed some people using this list elsewhere, please link to this thread, mostly to ensure people can stay up to date.

Fixed x1800.

 

Ultra High end
--------
7900GTX SLI = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (500/600) *2 SM3
7800GTX SLI = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTX = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (500/600) SM3
X1800 XT = 16 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (550/650) SM3
7800GTX = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (400/550) SM3
7900GS = 20 PSUs, 7 vertex units, 12 ROPs, 256 bit mem (375/500) SM3
X1800 = 12 PSUs, 8 vertex units, 12 ROPs, 256 bit mem (450/500) SM3

Mid-High end
--------
x800xt = 16 PSUs, 6 vertex units, 16 ROPs, 256 bit mem (480/550)
6800U = 12 PSUs, 5 vertex units, 16 ROPs, 256 bit mem (450/550) SM3
x800pro = 12 PSUs, 6 vertex units, 12 ROPs, 256 bit mem (400/400)
7800 = 16 PSUs, 6 vertex units, 16 ROPs, 256 bit mem (250/329) SM3
6800 = 12 PSUs, 5 vertex units, 12 ROPs, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 PSUs, 5 vertex units, 4 ROPs, 128 bit mem (470/470) SM3
7600 = 8 PSUs, 5 vertex units, 8 ROPs, 128 bit mem (~450/500) SM3
x700 = 8 PSUs, 6 vertex units, 8 ROPs, 128 bit mem (350/300)
6600 = 8 PSUs, 3 vertex units, 8 ROPs, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (432/400) SM3
x1300 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 PSUs, 3 vertex units, 4 ROPs, 64 bit mem (450/450) SM3
x600 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (400/250)
x600se = 4 PSUs, 2 vertex units, 4 ROPs,128bit mem (358/277)
9700 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (450/213)
9600 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (333/? (240 turbo))
6400 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (400/350) SM3
7300 = 4 PSUs, 3 vertex units, 4 ROPs, 64 bit mem (350/350) SM3
x300 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (300/230)
6200 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 PSUs, 1 vertex unit, 2 ROPs, system memory only (475/N/A)** SM3
6100 = 2 PSUs, 1 vertex unit, 2 ROPs, system memory only (425/N/A) ** SM3
xpress200 = 2 PSUs, 2? vertex units, 2 ROPs, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 PSUs, (Array) vertex units, 4 ROPs, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Ive noticed some people using this list elsewhere, please link to this thread, mostly to ensure people can stay up to date.

PSU = Pixel shader units
ROP = Raster Operations Units, takes the info from the Pixel shader about each pixel and assembles it into the final picture, this is where AA is added.

 

Ultra High end
--------
7900GTX SLI = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (500/600) *2 SM3
7800GTX SLI = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (400/550) *2 SM3

High end
--------
7900GTX = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (500/600) SM3
X1800 XT = 16 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (550/650) SM3
7800GTX = 24 PSUs, 8 vertex units, 16 ROPs, 256 bit mem (400/550) SM3
7900GS = 20 PSUs, 7 vertex units, 12 ROPs, 256 bit mem (375/500) SM3
X1800 = 12 PSUs, 8 vertex units, 12 ROPs, 256 bit mem (450/500) SM3

Mid-High end
--------
x800xt = 16 PSUs, 6 vertex units, 16 ROPs, 256 bit mem (480/550)
6800U = 12 PSUs, 5 vertex units, 16 ROPs, 256 bit mem (450/550) SM3
x800pro = 12 PSUs, 6 vertex units, 12 ROPs, 256 bit mem (400/400)
7800 = 16 PSUs, 6 vertex units, 16 ROPs, 256 bit mem (250/329) SM3
6800 = 12 PSUs, 5 vertex units, 12 ROPs, 256 bit mem (300/300) SM3

Mid range
--------
x1600 = 12 PSUs, 5 vertex units, 4 ROPs, 128 bit mem (470/470) SM3
7600 = 8 PSUs, 5 vertex units, 8 ROPs, 128 bit mem (~450/500) SM3
x700 = 8 PSUs, 6 vertex units, 8 ROPs, 128 bit mem (350/300)
6600 = 8 PSUs, 3 vertex units, 8 ROPs, 128 bit mem (300/300) SM3

Low end
--------
x1400 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (432/400) SM3
x1300 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (hyper mem) (400/325) SM3
7400 = 4 PSUs, 3 vertex units, 4 ROPs, 64 bit mem (450/450) SM3
x600 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (400/250)
x600se = 4 PSUs, 2 vertex units, 4 ROPs,128bit mem (358/277)
9700 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (450/213)
9600 = 4 PSUs, 2 vertex units, 4 ROPs, 128 bit mem (333/? (240 turbo))
6400 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (400/350) SM3
7300 = 4 PSUs, 3 vertex units, 4 ROPs, 64 bit mem (350/350) SM3
x300 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (300/230)
6200 = 4 PSUs, 2 vertex units, 4 ROPs, 64 bit mem (300/300) SM3

Ultra low end
--------
6150 = 2 PSUs, 1 vertex unit, 2 ROPs, system memory only (475/N/A)** SM3
6100 = 2 PSUs, 1 vertex unit, 2 ROPs, system memory only (425/N/A) ** SM3
xpress1150 = 2 PSUs, 2? vertex units, 2 ROPs, 128^ bit & 32* bit mem (400/ (SYS MEM & ?*))
xpress1100 = 2 PSUs, 2? vertex units, 2 ROPs, 128^ bit & 32* bit mem (300/ (SYS MEM & ?*))
xpress200 = 2 PSUs, 2? vertex units, 2 ROPs, 128^ bit & 32* bit mem (300/ (SYS MEM & 350*))
5x00 = 4 PSUs, (Array) vertex units, 4 ROPs, 128 bit/64 bit mem (%/%)
All other integrated solutions (GMA 900, 950 etc.)

*Dependant on memory attached to side port
** Performance of these two is identical
^system mem bus width
% Covers entire series, pretty awful due to high power consumption and generally poor arc, 5700go would rate a little higher but those above it perform much better in DX9 titles such as HL2 therefore they get stuck at the bottom.

Credit where credit is due, recently Neeyik ( http://www.neeyik.info/3dspecs/) has really helped get core details.

Ive noticed some people using this list elsewhere, please link to this thread, mostly to ensure people can stay up to date.

PSU = Pixel shader units
ROP = Raster Operations Units, takes the info from the Pixel shader about each pixel and assembles it into the final picture, this is where AA is added.

You know while the x1100/x1150/x200 can have memory attached, really DONT BOTHER.

 

Just to let you guys know this thread is going STICKY (my first ever) =D

Well I say this thread, actually I will create a new one! Its amazing to see how the list has evolved. Thanks for those who put in input and I hope it has been of use (and will continue to be).

 

Good job! Thanks.