x1600xt crossfire

[althes]

nice one sabre, saaya the cards wont go anywhere with the mods so I think it really is a 630 bios lock,
This is stupid in my opinion.

[sabrewolf732]

nice one sabre, saaya the cards wont go anywhere with the mods so I think it really is a 630 bios lock,
This is stupid in my opinion.

can you mod the bios? the 9500p had the same problem

[althes]

The cards are packed and going back.

[sabrewolf732]

The cards are packed and going back.

x1600xt's? you rmaing them w/ restocking fee? if so Ill buy one from ya lol

[vapb400]

Wow. I'm glad I didn't jump the gun on these.

[saaya]

well they are not bad cards... they beat a 6600gt, but the ocing lock sux...
and they dont scale with cpu power...
they are cheap, and will hopefully get even cheaper... but id rather recommend people to pay 2x the price and get a decent card.
wich lasts 2x as long and lets you play with high settings for at least 1 year...
6800gs, 1800xl, 7800gt... way better price performence deals if you ask me...

[STEvil]

[QUOTE=saaya]correct me if im wrong but dont all sites and people meassure the power consumption of the hardware, and not how much power flew through it?[quote]
If you dont measure the full power being used are you really measuring the correct "power" ?

only the differential between what flew in and what flew out gets measured afaik. so the 40W is what was brought to the card and didnt leave it, and the only way it can leave the card would be to ionize the air around it, wich would be 0.0000w id guess
or through heat.

heat or through to ground like any circuit.

where do they say the card draws all the power through the board?

They didnt on the page you linked, but they didnt state they were also measuring the power connector on that same page either (ie: I didnt read the whole article since all I wanted was that one page...).

as i said, they wanted to get as close to the 100% accuracy as possible, why would you go for something less accurate if its possible to get better results without a big effort?
if you have a question about how they did it just email them, they always replied to my emails so far...

My point is that the method they used didnt seem as accurate as could have been done.

dude, right n the very first paragraph of the page of the article i linked you to:

READ dude, READ!

didnt have time to read the whole thing, my bad.

O rly? :P
still dont get what you mean with
please explain

expect a x700 performence wise when you get your 1600 and bench it, otherwise you will be dissapointed
its a bit faster than a x700 at stock, quite a bit in some situations, but its not a x800 level card...

Seems better i just about everything than the x700 to me.. but i'm still using 9700pro

its .13 droop wich shows my psu cant keep up on the 12v rail.
correct me if im wrong, bu droop means the voltage drops because the draw is so big that the circuit becomes less efficient hence the voltage drops.
the .4v is just the circuit resistance, wich is imo pretty large.
for vdim its usually .2 and on this board theres a molex plug 1cm above the first pciE slot... so its kinda weird the traces have such a high resistance...

http://www.xtremesystems.org/forums/...0&postcount=94
12.00v to 11.93v is 0.07v.... (need to use load numbers)
40/11.93 = 3.35A (all on 12v rail? unknown..)
3.35A * 0.07v = 0.235w being generated within the traces of the motherboard or at the connector.

[althes]

x1600xt's? you rmaing them w/ restocking fee? if so Ill buy one from ya lol

No i am rmaing them because they arent working at all.
See if I can get back a better pair

[sabrewolf732]

No i am rmaing them because they arent working at all.
See if I can get back a better pair

you're rmaing them cause they dont oc well? that's shady

[althes]

NO they are not working at all wont boot up I am getting the beeps

[sabrewolf732]

NO they are not working at all wont boot up I am getting the beeps

oh that sucks hard Where you got them from?

[saaya]

stevil, i dont want to argue with you, please make your point instead of dissaproving what i said and what xbit said.
so you think they meassured the power consumption bad?
how? and please explain what you mean and dont post a half sentence :P
and you think the power consumption of a card =! the heat dissipation of it?
if so then please specify what makes you think so and how you could/can prove it.

They didnt on the page you linked, but they didnt state they were also measuring the power connector on that same page either (ie: I didnt read the whole article since all I wanted was that one page...).

didnt have time to read the whole thing, my bad.

in the first paragraph of the site i linked you to they clearly say they also manipulated the power plug to meassure the draw through it
http://www.xbitlabs.com/articles/vid...-x1000_14.html
To measure how much power the graphics accelerator consumes through the external connector, we used an adapter equipped with special shunt and connectors

My point is that the method they used didnt seem as accurate as could have been done.

? didnt you mention that there are other ways to meassure the power consumption you would have used wich are LESS acurate than what they did?
and for somebody who didnt even read the first paragraph of page 14 of a 30 page article you dont really have the right to criticize their methods if you ask me. :P
plus how would they have done it in a better way?
your alternative would have been even less accurate.

http://www.xtremesystems.org/forums/...0&postcount=94
12.00v to 11.93v is 0.07v.... (need to use load numbers)
40/11.93 = 3.35A (all on 12v rail? unknown..)
3.35A * 0.07v = 0.235w being generated within the traces of the motherboard or at the connector.

so whats your point?

why would the cards have a 630 clock limit... weird...
why 630?

and i dont think its that they dont get enough power through the mainboard... as i said, on my board theres a moley next to the pciE slots and i adjusted the 12v rail and theres no difference at all...

[STEvil]

As I said, I skimmed over it rather quickly...

My bad

.

Also as I said, their methods seemed to be less accurate than what could have been done, which was because I didnt take the time to read it thoroughly enough... I thought the shunt was installed between the PCI-E slot and the video card..

My point is that the method they used didnt seem as accurate as could have been done.

We used a special testbed based on a modified mainboard that allowed connecting special measuring devices to the 12V and 3.3V power lines leading to the PCI Express x16 slot. To measure how much power the graphics accelerator consumes through the external connector, we used an adapter equipped with special shunt and connectors

no +5v?

? didnt you mention that there are other ways to meassure the power consumption you would have used wich are LESS acurate than what they did?

There are two ways to do it, one more and one less accurate.

and for somebody who didnt even read the first paragraph of page 14 of a 30 page article you dont really have the right to criticize their methods if you ask me. :P

They didnt give us a picture of what they did, do they have the right to tell us they've done something without showing us as proof? Yes, i'm just being hard on them here...

plus how would they have done it in a better way?
your alternative would have been even less accurate.

The more accurate alternative would have been to connect the power meter right at the PCB of the PSU for each rail (+3.3v, +5v, +5vsb. +12v) and use a single rail psu (no split +12v since those could skew readings).

so whats your point?

my point is you are getting negligible losses at the motherboard/connector so your PSU and/or mobo/pci-e slot are likely not at fault. 0.07v is a far cry from 0.4v or 0.13v as well.

and please explain what you mean and dont post a half sentence :P
and you think the power consumption of a card =! the heat dissipation of it?
if so then please specify what makes you think so and how you could/can prove it.

There really is not much more to say than voltage/amperage when encountering resistance produces heat. This is a basic rule of electricity or any kind of energy. Energy encountering resistance must dissipate some of its value to continue to its destination.

As to proving that not all energy "consumed" goes directly to heat... first lets see you cool an opteron or X2 putting out 118w with the x1800xt cooler, then you can explain to me where you get that 118w from if there is no circuit for it to travel.

[sabrewolf732]

stevil, i dont want to argue with you, please make your point instead of dissaproving what i said and what xbit said.
so you think they meassured the power consumption bad?
how? and please explain what you mean and dont post a half sentence :P
and you think the power consumption of a card =! the heat dissipation of it?
if so then please specify what makes you think so and how you could/can prove it.

Power consumption != heat, thats just common sense. If electronics produced heat with all of its watts we wouldn't be typing right now.

[saaya]

stevil
pciE cards cant use the 5v rail, the pciE slot only provides 3.3v and 12v, and the power plug is 12v only, at least for the cards ive seen.

how is connecting to the psu pcb more accurate? then you meassure the resistance in the psu cables... plus you couldnt meassure what power goes to the videocard and what to the cpu and memory etc, this would only work to meassure the overal system power consumption.

and they did post pics when the first used this method of meassuring the power consumption of cards when r420 launched iirc... or when 6800 launched? yeah i think when 6800 launched.

a 1800xt consumes 118W, the card, not the vpu!
the 1800xt vpu is probably only around 70W id say, considering the large die that produces the heat, there is no problem cooling it with the rather small and simple heatsink.
and ive seen some guys put a geforce fx leafblower on an athlonxp and they got really nice temps, those heatsinks are pretty good since they have a lot of cfm and the air is flowing steadily over all fins and theres no dead spot in the center of the heatsink like with all "normal" heatsinks.

Power consumption != heat, thats just common sense. If electronics produced heat with all of its watts we wouldn't be typing right now.

afaik power consumption = heat...

[sabrewolf732]

If all the power is being transformed into heat, how would anything work? What would separate pc's from space heaters?

[saaya]

If all the power is being transformed into heat, how would anything work? What would separate pc's from space heaters?

space heater = free electrons of the electron flow (electricity) randomly hit other electrons bound to atoms, the atoms and electrons start moving faster.
heat=speed matter is moving at, the faster the matter moves the hotter it gets

petty much like a party where matter = people and electricity = booze
the more boze you induce to the party the more it will make people start to dance n freak out
the faster peopl dance and the more they freak out, the better the party = the bigger the heat.

the difference between a heater and an ic is that the electrons dont just hit random atoms but that they run through traces and shut switches on and off.
on the way they hit atoms and create heat though, and switching the transistors also results in heat afaik...

[cadaveca]

teh heat is generated when the transitors switch. electrical power turned to mechanical, then heat, via friction. this is the power the cpu/gpu comsumes. Heat is also given off in power leakage tho, so all heat given off by the cpu/gpu is not directly related to power consumption, as the leakage will vary from chip to chip.

[saaya]

well from what i know the leakage only determins how much of the power the ic uses is actually used to do some usefull work, aka switching a transistor, and how much goes wasted by just leaking out of the traces before it reached a transistor to switch.

afaik the leakage of prescott is pretty high with 40% or so.
so of the 140W+ a prescott consumes 56W get lost inside the traces before they can switch a transistor, and the rest, 84W get consumed/turned into heat by switching transistors.

[sabrewolf732]

space heater = free electrons of the electron flow (electricity) randomly hit other electrons bound to atoms, the atoms and electrons start moving faster.
heat=speed matter is moving at, the faster the matter moves the hotter it gets

petty much like a party where matter = people and electricity = booze
the more boze you induce to the party the more it will make people start to dance n freak out
the faster peopl dance and the more they freak out, the better the party = the bigger the heat.

the difference between a heater and an ic is that the electrons dont just hit random atoms but that they run through traces and shut switches on and off.
on the way they hit atoms and create heat though, and switching the transistors also results in heat afaik...

So if all the power being consumed is going directly to heat, where is the energy going needed to open and close the gates? You're trying to get something (computer output) for nothing (saying all power = heat)

[cadaveca]

well from what i know the leakage only determins how much of the power the ic uses is actually used to do some usefull work, aka switching a transistor, and how much goes wasted by just leaking out of the traces before it reached a transistor to switch.

afaik the leakage of prescott is pretty high with 40% or so.
so of the 140W+ a prescott consumes 56W get lost inside the traces before they can switch a transistor, and the rest, 84W get consumed/turned into heat by switching transistors.

uh, kinda, but no. those 56w get lost whether they are switching transitors or not. leakage can be different things, but generally, simply current leaking out of the circuit. becasue silicon is an insulator, it will turn this leakage into heat, hopefully enough of it before the leakage is enough to interfere with the switching of the transistor where the leakage lands. this sort of leakage is what affects the r520, to the point that the leakage creates a path to ground, and stops the flow of the circiut, leading to the gpu locking up @ higher voltages.

[saaya]

So if all the power being consumed is going directly to heat, where is the energy going needed to open and close the gates? You're trying to get something (computer output) for nothing (saying all power = heat)

no, the elelctrons carry power wich they use to hit the switches and change them from 0 to 1 or 1 to 0, then switches however dissipate the power as heat again.

if you hit them and set them to 1 they keep the energy until they get switched off, then the energy gets transformed to heat and they move to 0.
or its the other way around, dont remember.
you dont get anything for free, you had to dump loads and loads at electrons to make them move the switches. in the end the energy gets transformed to heat, but the switching of the transistors doesnt come for free at all.
the heat is just a bonus, wich you can use to heat your room or cook your food in theory
but the hea is more like a sideproduct.

somebody could as well say that by keeping cattle to produce methaine gas in a bio reactor you get meat for free...
or by keeping cattle to have meat you get methaine gas for free to power your house

uh, kinda, but no. those 56w get lost whether they are switching transitors or not. leakage can be different things, but generally, simply current leaking out of the circuit. becasue silicon is an insulator, it will turn this leakage into heat, hopefully enough of it before the leakage is enough to interfere with the switching of the transistor where the leakage lands. this sort of leakage is what affects the r520, to the point that the leakage creates a path to ground, and stops the flow of the circiut, leading to the gpu locking up @ higher voltages.

yeah, thats what i meant, the leakage is lost anyways, they wont be able to switch transistors because they get lost along the way.
but your right, they can still affect the ic.
however, i doubt that the lockups of a r520 or any other ic are caused by a shortage within the ic.
a shortage within the ic would kill it i think.
it would be like a flash, a big load of current would travel through the path where electrons managed to escape the traces and go to ground, and the traces they flow through would literally melt.

[saaya]

LOL, i tried all 3 ati bios editors i could find but none of them will work, 2say bad file and 1 crashes

so i opened the bios file i saved from my card with notepad and guess what the first line says, lol

RV530 PCI_EXPRESS DDR2 A67611 RV530PRO TEST BIOS DDR2 500e/400m

test bios??? on a retail card? lol...

and this is even more interesting:

YOU HAVE NOT CONNECTED THE POWER CABLE TO YOUR VIDEO CARD.PLEASE REFER TO THE 'GETTING STARTED GUIDE' FOR PROPER HARDWARE INSTALLATION.

so those cards had a power plug originally?
the bios is from 31st of oktober.
this explains why the cards suck so hard on the 12v rail, originally they had a power plug i think, and the power plug is 12v only.
but ati wanted them without power plug i guess...

or im just wrong and es cards always have a power plug, and thats why theres this message in the bios.

[cadaveca]

however, i doubt that the lockups of a r520 or any other ic are caused by a shortage within the ic.
a shortage within the ic would kill it i think.
it would be like a flash, a big load of current would travel through the path where electrons managed to escape the traces and go to ground, and the traces they flow through would literally melt.

you'd be right, but you are not. the leakage is enough that the GPU cannot continue to run...and the issue is a "soft ground", something that is kinda built in to deal with leakage issues, but way so bad at this point that the gpu would falter...

And yes, it IS melting these R520 cores. i think the issue IS causing definate damage, as each time i run the card over 1.275, i get more and more artifacts @ stock.

[saaya]

hmmm over 1.275 already gives you artifacts at stock?
i thougt people were running 1.5v with decent cooling?
are you on air?
how high did you set vgpu?