x1600xt Out now! any one OC it?

I just scored 6335 3dmark05 with my 2.45ghz athlon xp mobile, with my x800xt AIW at 550/580

so this overclocked x1600xt is pretty darn close

Wow the x1600xt looks like a card that can hang with the boys

Quote:

---

Originally Posted by althes

Wow the x1600xt looks like a card that can hang with the boys

---

does that make it a pedophile?

Quote:

---

Originally Posted by Cybercat

does that make it a pedophile?

---

:p:
;)

saaaaaya where you at homie ? :slap:

Quote:

---

Originally Posted by Cybercat

does that make it a pedophile?

---

That statement is just so wrong:D

Lets keep away from those type of comments ;) But yea, I'm as interested as any on how to increase volts on these. Its a feature that was guaranteed as a part of the architecture...

Everyone with any type of x1600 seriously needs to follow x1800 ocing rituals to get results imo.

Perkam

What are they pray tell

http://www.newegg.com/Product/Produc...82E16814102656

is that really what i think it is?

that's totally the wrong set of pictures.

where is saaya? must be having one hell of a time with those x1600xt's ;)

Word around the Net is that any x1600 will work in CF without having a CF Edition master card. That much I've confirmed, but I was curious if anyone knew if that was limited only to the ATI chipset mobos or if that goes for the 955X as well?

I am trying to decide between x1600XTs in CF or a single x1800XL. The 1800 seems to be winning the toss up anyway, but if the 1600s in CF require a new mobo, that will just seal the deal.

Any info is appreciated!!

yup you need a CF mobo but Inqurier tested two X1600XT and passed a X1800XT by a lil bit.Well I am in Canada and one X1600XT is $195x2=$390 and a ASUS A8R-MVP(best CF board availabe) $125 so a grand total of $515. The cheapest X1800XT is $649 so CF is a nice soultion :D

Nice Find...though those may be the wrong pictures, whoever is looking forward to a CF X1600XT solution should grab one of those for higher performance than with two regular XTs.

Perkam

I figured that was just a gimmick to make it sound more appealing. X1600 series don't need master cards last I heard. Notice it says "Crossfire supported", not Crossfire Edition. In fact every one of Sapphire's X1600XT cards say that.

i guess saaya did not get his card today? hopeing to see some results soon :D

Quote:

---

Originally Posted by Cybercat

what it is is the four in 4x3 is the four fragment pipes in each quad. There are three of them.

I think everyone took the claims from ATI that each pipe is three times as efficient, in tandem with the confusion about TMUs versus pipelines, to mean that each fragment is somehow really three fragments, or some messed up explaination like that. It's a big mess, and not a lot of people seem to really get it straight. You can't have 16 fragments in a single quad, since obviously a quad implies only four.

Let me try to clear this up right now. It's 12 shader fragments, or three quads, four texture mapping units, four ROPs. The texture units in the RV530 are sort of enhanced from those in the R520. Each texture mapping unit can serve three fragments at once. So it only takes four to handle all shader units in the RV530. Beyond3D talks about this as well.

For the R580, there are 48 shader fragments, or 12 quads, 16 texture mapping units, and 16 ROPs. Again, each texture mapping unit can address up to three fragments at once, so 16 covers all 48 shader units.

For the rumored RV560's configuration, there 24 shader fragments, which means there must be 8 TMUs to cover them all. The number of ROPs doesn't matter, since there's traditionally a crossbar between them all, but assumably they also work in quads, just like the shader units.

---

hmmmm this makes sense, but its confusing me as what i heard and read before was totally different... what about pixelprocessor?
what i heard was that each pixelpileline not contains 3 pixel processors, so its like a tripply pipeline. is this what people misunderstood and ati actually only said the pixel pileines are 3x as efficient?

the thing about the texture units covering all the pixe pipelines is what ati claims though, and if you ask me, and if you ask the reviewers and look at the benchmarks, it doesnt look like they are as efficient as ati claims.

Quote:

---

Originally Posted by StixxX

yup you need a CF mobo but Inqurier tested two X1600XT and passed a X1800XT by a lil bit.Well I am in Canada and one X1600XT is $195x2=$390 and a ASUS A8R-MVP(best CF board availabe) $125 so a grand total of $515. The cheapest X1800XT is $649 so CF is a nice soultion :D

---

no offense, but theinq knows sht about benchmarking :D
two 1600xts beat a 1800xl in crossfire, but only in 2k3 and 2k5, in games its a totally different picture!
the 1800xl will kill the 1600xts in crossfire if you ask me... we havent seen any benches yet so who knows... but from what ive seen from 1800s and 2850s in crossfire they scale only 10-25% in games over a single card, its only in 2k3 and 2k5 where they scale 40% and higher.

so DONT get 1600 in crossfire, or wait and see what everybody can get out of those cards in cf :D
in a few days you will know :)

Quote:

---

Originally Posted by Welz

i guess saaya did not get his card today? hopeing to see some results soon :D

---

still didnt get it! :mad:

oh and that card opn newegg... if the pics are not a mix up it looksd like sapphire put 1600s on 1800 pcbs :slobber: those would be very nice cards, should oc really nice!
somebody call newegg and check if the pics are correct!
165$ damn... i wish 1600xt cards would be that cheap over here... they cost almost 25% more over here

saaya, your top quote is all messed up. :p:

EDIT:

Quote:

---

Originally Posted by saaya

hmmmm this makes sense, but its confusing me as what i heard and read before was totally different... what about pixelprocessor?
what i heard was that each pixelpileline not contains 3 pixel processors, so its like a tripply pipeline. is this what people misunderstood and ati actually only said the pixel pileines are 3x as efficient?

---

I've never heard of a modern shader architecture from NVIDIA or ATI yet using pipelines in sets of three.

Quote:

---

Originally Posted by saaya

the thing about the texture units covering all the pixe pipelines is what ati claims though, and if you ask me, and if you ask the reviewers and look at the benchmarks, it doesnt look like they are as efficient as ati claims.

---

Well, you're right. Having four texture units means it has the texturing fillrate of only four units, no better than an X1300, and MUCH less than the X700 or mainstream X800 cards. ATI was right to make the call that games demand more shader power than texturing power, as that's certainly where we're going now, but four simply isn't enough. The "enhanced" part of their new TMUs simply means that they're, I guess you could say, multithreaded, to accept more than one pipe. However, just because they can accept them doesn't mean they're as fast as 12 true texture units. Their capacity increased, but not their performance. Hope that makes sense.

Quote:

---

yup you need a CF mobo but Inqurier tested two X1600XT and passed a X1800XT by a lil bit.Well I am in Canada and one X1600XT is $195x2=$390 and a ASUS A8R-MVP(best CF board availabe) $125 so a grand total of $515. The cheapest X1800XT is $649 so CF is a nice soultion

---

Dang, I was hoping it would just beat an 1800XL let alone an 1800XT. Rats that I'd need an ATI mobo to do it :( The 955 chipset supports CF, but I guess that the CF 1600XTs won't work without a master card in that case...damn. Thanks for the info tho!

Quote:

---

Originally Posted by Cybercat

saaya, your top quote is all messed up. :p:

EDIT:


I've never heard of a modern shader architecture from NVIDIA or ATI yet using pipelines in sets of three.


Well, you're right. Having four texture units means it has the texturing fillrate of only four units, no better than an X1300, and MUCH less than the X700 or mainstream X800 cards. ATI was right to make the call that games demand more shader power than texturing power, as that's certainly where we're going now, but four simply isn't enough. The "enhanced" part of their new TMUs simply means that they're, I guess you could say, multithreaded, to accept more than one pipe. However, just because they can accept them doesn't mean they're as fast as 12 true texture units. Their capacity increased, but not their performance. Hope that makes sense.

---

Good info.

Quote:

---

Originally Posted by Cybercat

saaya, your top quote is all messed up. :p:

EDIT:


I've never heard of a modern shader architecture from NVIDIA or ATI yet using pipelines in sets of three.


Well, you're right. Having four texture units means it has the texturing fillrate of only four units, no better than an X1300, and MUCH less than the X700 or mainstream X800 cards. ATI was right to make the call that games demand more shader power than texturing power, as that's certainly where we're going now, but four simply isn't enough. The "enhanced" part of their new TMUs simply means that they're, I guess you could say, multithreaded, to accept more than one pipe. However, just because they can accept them doesn't mean they're as fast as 12 true texture units. Their capacity increased, but not their performance. Hope that makes sense.

---

best way to check this is to bench both cards (x1300/x1600) at same freq in shadermark2.0:stick:

i can test x1300 let say at 600/400 and cpu rig set to 2400mhz
and someone with x1600 will provide scores from close to this system

then we will see what is worth ati 12 shader 4 pixel architecture

anyone ready to test this theory??

You know what to expect. The X1600 will perform better in a shader benchmark. Afterall it has theoretically three times the shader hardware available than the X1300.

If you really want to test my statement, actually use a benchmark that tests the area of which I'm talking about: texturing. Check the texture fillrates with 3DMark01, for example. That will prove what I, ATI, and any other website that's reviewed and analyzed the two cards have been saying: they both have the same amount of texture units.

Quote:

---

Originally Posted by Cybercat

saaya, your top quote is all messed up. :p:

---

yeah sorry, fixed it :)

btw cards still didnt arrive :upset:

Cybercat, what i heard was that there are 4 pixel pipelines, a quad, but each pipeline has 3 pixel processors. pretty much like a tripple pipeline or something like that. you know g7x have 2 pixel processors in their pipeline. well i dont think you call them pixel processors, but you know what i mean.. i hope :D
they have two parts in the pipeline to process shader code, thats at least the way i remember it.

Quote:

---

Originally Posted by Cybercat

Well, you're right. Having four texture units means it has the texturing fillrate of only four units, no better than an X1300, and MUCH less than the X700 or mainstream X800 cards. ATI was right to make the call that games demand more shader power than texturing power, as that's certainly where we're going now, but four simply isn't enough. The "enhanced" part of their new TMUs simply means that they're, I guess you could say, multithreaded, to accept more than one pipe. However, just because they can accept them doesn't mean they're as fast as 12 true texture units. Their capacity increased, but not their performance. Hope that makes sense.

---

i get what you mean, but i dont agree... shader power is more important, but they cut off too much textureing power if you ask me... do the tmus also affect bump mapping and "normal maps" or however they were called?

Quote:

---

Originally Posted by boostedevo

Dang, I was hoping it would just beat an 1800XL let alone an 1800XT. Rats that I'd need an ATI mobo to do it :( The 955 chipset supports CF, but I guess that the CF 1600XTs won't work without a master card in that case...damn. Thanks for the info tho!

---

cf will work without a mastercard in the 955x chipset as well... there are no 1600 and 1300 master cards...

Quote:

---

Originally Posted by Lightman

best way to check this is to bench both cards (x1300/x1600) at same freq in shadermark2.0:stick:

i can test x1300 let say at 600/400 and cpu rig set to 2400mhz
and someone with x1600 will provide scores from close to this system

then we will see what is worth ati 12 shader 4 pixel architecture

anyone ready to test this theory??

---

if i had my 1600s id do it... hope i finally get them tomorrow... :/

Quote:

---

Originally Posted by Cybercat

You know what to expect. The X1600 will perform better in a shader benchmark. Afterall it has theoretically three times the shader hardware available than the X1300.

If you really want to test my statement, actually use a benchmark that tests the area of which I'm talking about: texturing. Check the texture fillrates with 3DMark01, for example. That will prove what I, ATI, and any other website that's reviewed and analyzed the two cards have been saying: they both have the same amount of texture units.

---

yeah, but would be really interesting to compare the 1300 and 1600 architecture at the same clocks... i wish i had my 1600s :(

I will dl shadermark 2.1 and test the x1600 in xfire an single card

Quote:

---

Originally Posted by saaya

no offense, but theinq knows sht about benchmarking :D
two 1600xts beat a 1800xl in crossfire, but only in 2k3 and 2k5, in games its a totally different picture!
the 1800xl will kill the 1600xts in crossfire if you ask me... we havent seen any benches yet so who knows... but from what ive seen from 1800s and 2850s in crossfire they scale only 10-25% in games over a single card, its only in 2k3 and 2k5 where they scale 40% and higher.

so DONT get 1600 in crossfire, or wait and see what everybody can get out of those cards in cf :D
in a few days you will know :)


still didnt get it! :mad:

oh and that card opn newegg... if the pics are not a mix up it looksd like sapphire put 1600s on 1800 pcbs :slobber: those would be very nice cards, should oc really nice!
somebody call newegg and check if the pics are correct!
165$ damn... i wish 1600xt cards would be that cheap over here... they cost almost 25% more over here

---

you've seen quite wrong then, as another forum member said the exact same thing yesterday. I showed him benchmarks of bf2, fear, quake 4 and other games where the game was more to the effect of 70%, sometimes close to 100.