Anand's x1950p review: CrossFire Done Right

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
Originally posted by: apoppin
Originally posted by: Creig
So apparently all the screaming and hysteria boiled down to a difference of 1-4 fps.





Wow.




(*yawn*)

no that wasn't the point . . . Derek Wilson ACCUSED ati of DISHONESTLY lowering the products specs AFTER it shipped.

he is wrong

that is all

time to move along


and Derek Wilson is a decent reviewer - usually. He just isn't of Anand's thoroughness and he lets his bias slip.
ATI sent AnandTech a X1900GT, they reviewed it, said <whatever they said> about the card.

ATI then changed the specs of the card. Now when people go to research the X1900GT, it's a different card than what AnandTech reviewed, but ATi is still calling it a X1900GT. It invalidates everything AT wrote using the X1900XT up until now, because ATi changed it.

And yes, they changed it after it launched. Look at Newegg, there are both the old and new styles in stock.

I'd call that dishonest.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: ViRGE
Originally posted by: apoppin
Originally posted by: Creig
So apparently all the screaming and hysteria boiled down to a difference of 1-4 fps.





Wow.




(*yawn*)

no that wasn't the point . . . Derek Wilson ACCUSED ati of DISHONESTLY lowering the products specs AFTER it shipped.

he is wrong

that is all

time to move along


and Derek Wilson is a decent reviewer - usually. He just isn't of Anand's thoroughness and he lets his bias slip.
ATI sent AnandTech a X1900GT, they reviewed it, said <whatever they said> about the card.

ATI then changed the specs of the card. Now when people go to research the X1900GT, it's a different card than what AnandTech reviewed, but ATi is still calling it a X1900GT. It invalidates everything AT wrote using the X1900XT up until now, because ATi changed it.

And yes, they changed it after it launched. Look at Newegg, there are both the old and new styles in stock.

I'd call that dishonest.
ati did NOT "hide" this as Derek says . . . look at Beyond3d:
Originally posted by: Pantalaimon
This was posted on Beyond3D on September 7th. It's still there on their news page towards the end of the page.

Beyond3D has learned of an upcoming specifications change to X1900 GT. The new spec lowers the clock speed by 63MHz, from 575MHz to 512MHz, while increasing the memory speed by 60MHz from 600MHz to 660MHz (1320MHz effective). A PCB redesign is also included to accomodate a fan change anticpated to produce improved accoustics, and HDCP capability is included by default on the new boards.

Damn sloppy of Derek Wilson/AT for not knowing.

and you can look at the box and determine which version you want - faster core or faster memory . . . for ~1% performance difference.

yeah . . . BIg deal

 

DerekWilson

Platinum Member
Feb 10, 2003
2,920
32
81
The point is not that there is going to be a performance difference in the games we currently test. The problem is that ATI lowered the *specifications* on a shipping product AT ALL.

This is not appropriate.

My assertion of dishonesty does not rely on their disclosure of what they are doing, but the fact that they are doing it. Not all consumers keep up with what ATI tells their partners or what is issued in press releases. Just because they openly and publicly acknowledge that they are doing something doesn't make it a good (or honest) thing to do.

Just because the performance difference is small doesn't excuse the fact that this product doesn't perform exactly as the original.

Current games aren't an absolute measure of performance. Future games which make increasing use of shader power could be more heavily impacted by core clock speed than current titles.

ATI has seen fit to change the name of a products for much smaller differences (as in the case of the X1300 XT and the X1650 Pro).

All of these things are true whether I test the product or not.

We knew about the Beyond3D story, but it would have been nice if ATI hadn't followed through on it. It's best to wait to comment on something like this until its actually happened, and that's what we did.

Our marketing contact at ATI didn't even know about the X1900 GT spec change until this week when I asked for a comment on it. This tells me that this isn't something ATI wanted to get out there.

If NVIDIA, Intel, AMD, S3, or anyone else did the same thing, I would be just as upset.

Everyone has a bias, and you should be very wary of people who would deny this. My bias is really hardware based. I tend to get excited about the potential of hardware, where we have to be very careful to report on the real world impact of hardware. Because my background is in computer engineering, I get really into the details. I really want "better hardware" regardless of price or real world feasibility to succeed. Case in point: I loved my PS2 linux kit. I don't care that it was hard to program for, PS2 is still my favorite console. And it probably will be until PS3 comes out.

The problem is that hardware with higher potential, more capabilities or whatever is not always the best value to the consumer. The hardware that lends itself best to game designers and that hits the best price target is really what AT calls better. Really, what we look for at AT is the best *engineered* hardware. Guys like Beyond3D look for the best hardware regardless of its usefulness.

I do my best to keep my opinion to myself when it isn't in line with what I should be recommending for our readers.

But I don't think anything about anyone changing the specs on any product by any degree (no matter how small) is alright. And I don't think this has anything to do with opinion or bias.

I'm certainly open to further discussion on the subject, and I'd love to hear feedback on this response.

::EDIT:: also, not all boxes display clock speeds ...
 

DerekWilson

Platinum Member
Feb 10, 2003
2,920
32
81
Originally posted by: crazydingo
The change in specs for the X1900GT was made because it was the best way to go in not decreasing the performance by much while having much better gpu temps. All this was informed to the AIBs more than a month back. So tying the change of spec in X1900GT to X1950 Pro's launch makes one look like a conspiracy theorist.

Derek got it wrong or maybe he isnt as informed as other reviewing authors. :laugh:

One more thing -- I got the information about why ATI changed the clock speed (dwindling 575MHz binned R580 supply) from ATI itself on Monday. It wasn't because of temps, it was because they are trying to get rid of old product they couldn't sell otherwise. This is fact, regardless of what other reviewers are saying.
 

redbox

Golden Member
Nov 12, 2005
1,021
0
0
Originally posted by: DerekWilson
The point is not that there is going to be a performance difference in the games we currently test. The problem is that ATI lowered the *specifications* on a shipping product AT ALL.

This is not appropriate.

My assertion of dishonesty does not rely on their disclosure of what they are doing, but the fact that they are doing it. Not all consumers keep up with what ATI tells their partners or what is issued in press releases. Just because they openly and publicly acknowledge that they are doing something doesn't make it a good (or honest) thing to do.

Just because the performance difference is small doesn't excuse the fact that this product doesn't perform exactly as the original.

Current games aren't an absolute measure of performance. Future games which make increasing use of shader power could be more heavily impacted by core clock speed than current titles.

ATI has seen fit to change the name of a products for much smaller differences (as in the case of the X1300 XT and the X1650 Pro).

All of these things are true whether I test the product or not.

We knew about the Beyond3D story, but it would have been nice if ATI hadn't followed through on it. It's best to wait to comment on something like this until its actually happened, and that's what we did.

Our marketing contact at ATI didn't even know about the X1900 GT spec change until this week when I asked for a comment on it. This tells me that this isn't something ATI wanted to get out there.

If NVIDIA, Intel, AMD, S3, or anyone else did the same thing, I would be just as upset.

Everyone has a bias, and you should be very wary of people who would deny this. My bias is really hardware based. I tend to get excited about the potential of hardware, where we have to be very careful to report on the real world impact of hardware. Because my background is in computer engineering, I get really into the details. I really want "better hardware" regardless of price or real world feasibility to succeed. Case in point: I loved my PS2 linux kit. I don't care that it was hard to program for, PS2 is still my favorite console. And it probably will be until PS3 comes out.

The problem is that hardware with higher potential, more capabilities or whatever is not always the best value to the consumer. The hardware that lends itself best to game designers and that hits the best price target is really what AT calls better. Really, what we look for at AT is the best *engineered* hardware. Guys like Beyond3D look for the best hardware regardless of its usefulness.

I do my best to keep my opinion to myself when it isn't in line with what I should be recommending for our readers.

But I don't think anything about anyone changing the specs on any product by any degree (no matter how small) is alright. And I don't think this has anything to do with opinion or bias.

I'm certainly open to further discussion on the subject, and I'd love to hear feedback on this response.

::EDIT:: also, not all boxes display clock speeds ...


Thanks for your response. While I can see why AT wanted to report on the change I really don't see why so many people are getting their panties in a bunch over it. Is it underhanded? In my opinion a little bit. The problem I have is when I buy a product I buy it for it's performance and IQ not it's specs. If I still get x amount of frames at x IQ then I don't care how ATI gets there. The specs on the box are really just to inflate your e-penis. What matters in the end is what you do with those specs.

I don't think such a small change should warrant a new branding, and judging from your past article on the x1650pro I am not sure you do either:
Quote from Article:
The X1650 Pro is identical to the X1600 XT except for a 10MHz increase in core clock and memory clock frequency. Yes, an entirely new product was created out of a 10MHz bump in GPU/memory clocks.

So what would you want ATI to do? They can't please you whether they rename the card or not. They're kind of in a damned if they do, damned if they don't position aren't they?

 

Slammy1

Platinum Member
Apr 8, 2003
2,112
0
76
You know, there's a big difference between a 10MHz/10MHz jump and a 63MHz/120MHz change, especially when 1 is a "new" product the other is a shipping product. Hardly damning evidence of bias.
 

hmorphone

Senior member
Oct 14, 2005
345
0
0
I take it that someone missed these two threads?

x80064's thread

My thread

Look at the dates and the number of replys, and then go look at Beyond3D on September 7th

It's also interesting to see some folks praising the X1950pro and slamming the X1900GT in the same breath, when the basic specs for the cards are identical. I know the speeds vary somewhat between the 3 cards, but I would imagine my X1900GT rev 1 at 650/1425 would whip a stock X1950pro. The overclocks for the X1950pro are as of yet not fully known as Overdrive seems to be the only thing that supports it at the moment, and Overdrive is somewhat limited. To a less extent this also applies to the X1900GT rev 2 as ATItool and such don't yet work well with it either. A more important point may be that you can't change voltages in software with the X1900GT rev 2 and probably can't with the X1950pro either. I do like the improvements in the die shrink and crossfire, but why slam an under $200 X1900GT and praise the under $200 X1950pro this much? I will say though that they should've put rev 2 on the X1900GT rev 2 box or something.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
thanks for replying in thread, Mr Wilson and explaining why you wrote what you did.

i think you are did make a claim of 'dishonesty' that is not well founded

i edited my topics title again and throw it back to the forum again
 

Roguestar

Diamond Member
Aug 29, 2006
6,045
0
0
So are you just going to change the topic's title any time someone you insult calls you on it, or are you going to make up your mind what your opinion is?
 

CaiNaM

Diamond Member
Oct 26, 2000
3,718
0
0
Originally posted by: DerekWilsonJust because they openly and publicly acknowledge that they are doing something doesn't make it a good (or honest) thing to do.

yet openly and publicly acknowledging what they are doing makes them "dishonest"? i could go on and argue against various other points, but for the sake of brevity...

6 of one, half dozen of the other.

Mr. Wilson, i certainly appreciate your participation in this thread, however i respectfully disagree with what you've done here, or at the very least the manner in which you have done it.

unless it can be shown that the customer is really getting less performance, image quality, or features than the "reasonable expectation" ati created from their statements, marketing, labeling, etc., you are making a "mountain out of a molehill".

one might even go further and suggest you are guilty of poor or even irresponsible journalism - after all the article is a technical review (about another product entirely), not an opinon piece or editorial which would be a more suitable format to share what seems to be your strong personal opinion on this subject. of course placing this in a launch review of a new product will certainly bring much more attention to this.. the word "sensationalism" comes to mind.

it's something i would expect of the Inquirer, but certainly not Anandtech....



 

Elfear

Diamond Member
May 30, 2004
7,126
738
126
Originally posted by: CaiNaM
Originally posted by: DerekWilsonJust because they openly and publicly acknowledge that they are doing something doesn't make it a good (or honest) thing to do.

yet openly and publicly acknowledging what they are doing makes them "dishonest"? i could go on and argue against various other points, but for the sake of brevity...

6 of one, half dozen of the other.

Mr. Wilson, i certainly appreciate your participation in this thread, however i respectfully disagree with what you've done here, or at the very least the manner in which you have done it.

unless it can be shown that the customer is really getting less performance, image quality, or features than the "reasonable expectation" ati created from their statements, marketing, labeling, etc., you are making a "mountain out of a molehill".

one might even go further and suggest you are guilty of poor or even irresponsible journalism - after all the article is a technical review (about another product entirely), not an opinon piece or editorial which would be a more suitable format to share what seems to be your strong personal opinion on this subject. of course placing this in a launch review of a new product will certainly bring much more attention to this.. the word "sensationalism" comes to mind.

it's something i would expect of the Inquirer, but certainly not Anandtech....

QFT
 

dreddfunk

Senior member
Jun 30, 2005
358
0
0
When I read the article, I personally didn't feel that Derek was making a 'mountain out of a molehill'. I knew from my experience of the AT forums, however, that the forums themselves were going to do just that. Green Team/Red Team -- everyone with a serious bias was going to come out of the woodwork to either jump on, or defend ATI depending on their personal inclination.

I agree with Derek: this isn't a very good precedent for any hardware company to set, full disclosure or no disclosure. Yes, I've read previous reviews where AT takes the various companies to task for having too many models and there is some, very minor, hypocrisy going on. Either the various model lines (x1900gt, xt, xtx, 7900gs, gt, gto, gtx) have a decent amount of wiggle room in their specs or they don't. If they do, then you'll get some decent performance fluctuations and you'll have to pay attention to what you're really getting. If they don't, then you're going to end up with a ridiculous number of models as each side continually tries to one-up the other at each and every price point, which is the current situation.

Personally, because I actually keep track of this stuff, I'd prefer fewer models with more wiggle room. Cars are like that: many times the car model doesn't absolutely specify, for example, what is really under the hood. You have to pay a bit of attention, which is fine by me.

My only real concern is the possibility that specs will be severely downgraded without disclosure. The case in question is a small downgrade currently, that could have a bigger impact on future games. My own opinion is that, if ATI or nVidia did this once in a blue moon and performance was impacted by less that 5% or so, then I wouldn't have a problem with it.

If and when it becomes a recurring theme for either ATI or nVidia and the performance impacts start to jump to 10% or more, I'll be ticked unless a better justification can be found than simply, "we miscalculated our ability to produce components that could perform up to the original specs," which is what ATI's admission of having fewer than required cores binning at full specs amounts to.

Just my opinion.
 

Creig

Diamond Member
Oct 9, 1999
5,170
13
81
Originally posted by: Derek Wilson
Just because the performance difference is small doesn't excuse the fact that this product doesn't perform exactly as the original.

So which would you rather see? ATI having two X1900GTs with nearly identical performance giving us yet another model number to add to their lineup? Or simply consolidate the two X1900GTs under one model number since the performance difference between the two is nearly non-existent?

You can't have it both ways.
 

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
Originally posted by: Creig
Originally posted by: Derek Wilson
Just because the performance difference is small doesn't excuse the fact that this product doesn't perform exactly as the original.

So which would you rather see? ATI having two X1900GTs with nearly identical performance giving us yet another model number to add to their lineup? Or simply consolidate the two X1900GTs under one model number since the performance difference between the two is nearly non-existent?

You can't have it both ways.
Option 3, discontinue the old X1900GT, don't release the new one.
 

xtknight

Elite Member
Oct 15, 2004
12,974
0
71
Yeah, changing specs after release = :thumbsdown:. Some of the birds chirping at Inquirer (as they so eloquently like to put it) may have heard of the change but when someone buys an X1900GT they don't know what they're getting. It could have been worse (the performance difference) though.
 

jiffylube1024

Diamond Member
Feb 17, 2002
7,430
0
71
Originally posted by: ViRGE
Option 3, discontinue the old X1900GT, don't release the new one.[/quote]

Then what does ATI do with all its R580 cores with either a broken quad or the inability to run at X1900XT/XTX clockspeeds? That's a big waste...
 

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
Originally posted by: jiffylube1024
Originally posted by: ViRGE

Option 3, discontinue the old X1900GT, don't release the new one.

Then what does ATI do with all its R580 cores with either a broken quad or the inability to run at X1900XT/XTX clockspeeds? That's a big waste...
Trash them, try to not screw up next time. It doesn't take a rocket scientist to know you'll have low-performing parts, and to plan accordingly. Releasing the X1950Pro when there's still a supply of R580's to go through is poor planning.
 

ronnn

Diamond Member
May 22, 2003
3,918
0
71
Originally posted by: ViRGE
Originally posted by: jiffylube1024
Originally posted by: ViRGE

Option 3, discontinue the old X1900GT, don't release the new one.

Then what does ATI do with all its R580 cores with either a broken quad or the inability to run at X1900XT/XTX clockspeeds? That's a big waste...
Trash them, try to not screw up next time. It doesn't take a rocket scientist to know you'll have low-performing parts, and to plan accordingly. Releasing the X1950Pro when there's still a supply of R580's to go through is poor planning.


Like who cares about ati's planning or nv's gs gto gt gtx, varied overclocked or stock speeds. If either company has extra gpu's they want to sell at very good prices - someone will buy them. I think maybe Derek would not have much else to write about if he started nit picking about every 2% change in performance. I am hoping to see quotes where he has taken nvidia to task for a false 2%, than at least this charge of being anti ati will be dispensed with. Many driver releases with small IQ differences eat up more than 2%. so he should have had many opportunities.
 

Munky

Diamond Member
Feb 5, 2005
9,372
0
76
Originally posted by: ViRGE
Originally posted by: jiffylube1024
Originally posted by: ViRGE

Option 3, discontinue the old X1900GT, don't release the new one.

Then what does ATI do with all its R580 cores with either a broken quad or the inability to run at X1900XT/XTX clockspeeds? That's a big waste...
Trash them, try to not screw up next time. It doesn't take a rocket scientist to know you'll have low-performing parts, and to plan accordingly. Releasing the X1950Pro when there's still a supply of R580's to go through is poor planning.

The x1950pro gives Ati bigger profit margins than the x1900gt, because it's a new core, on a smaller process and with only 36 pixel shaders to begin with. There's no reason to delay its launch just to get rid of all defective r580 cores. IMO, Ati could have priced the x1950pro at $250, because the $200 7900gs is overall significantly slower, and is hardly competitive with the x1900pro. Then to get rid of the r580 cores which do not meet the original x1900gt specs, they could have introduced a new $150 card with only 2 quads enabled, with 8 TU's and 24 PS. Such a card would go up against the 7600gt, which to this day has no modern competition from Ati.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: Roguestar
So are you just going to change the topic's title any time someone you insult calls you on it, or are you going to make up your mind what your opinion is?

i plan to change the title everytime it needs an update.

it needed an update when Mr Wilson's accusations proved to be overblown . . . and again to let the posters know he answered in the thread.

my opinion only changed once - when the facts became known

what ATi did was not ideal - certainly there is a lot of 'precedent' - but certainly NOT "underhanded" . . . nor did they attempt to "hide" anything

and Virge . . . i'm sorry but your suggestion to throw away the damaged cores is silly and doesn't make any business sense for any CPU/GPU company.
 

ViRGE

Elite Member, Moderator Emeritus
Oct 9, 1999
31,516
167
106
Originally posted by: apoppin
and Virge . . . i'm sorry but your suggestion to throw away the damaged cores is silly and doesn't make any business sense for any CPU/GPU company.
I know it doesn't make any business sense. I'm just sick and tired of there being so many products that are just the same thing over and over again. The speed binning has gotten out of hand, I yearn for the simpler days where there were a couple products per core max.
 

Elfear

Diamond Member
May 30, 2004
7,126
738
126
Originally posted by: ViRGE
Originally posted by: apoppin
and Virge . . . i'm sorry but your suggestion to throw away the damaged cores is silly and doesn't make any business sense for any CPU/GPU company.
I know it doesn't make any business sense. I'm just sick and tired of there being so many products that are just the same thing over and over again. The speed binning has gotten out of hand, I yearn for the simpler days where there were a couple products per core max.


Well if one of the byproducts of ATI selling unused cores (i.e. those that didn't make the grade) was to lower the price on the X1900XT(X) cards for awhile than I say more power to them. Whatever keeps both companies around to drive competition up is fine by me (as long as it's not underhanded of course).
 

Nelsieus

Senior member
Mar 11, 2006
330
0
0
Derek reported on ATI changing the specs of a GPU. I'm not sure why you wouldn't expect him to do so?

And no matter how small that change in spec was, it's the principle, imo, that counts. I have no problem with either ATI or nVidia changing the specs of any of their cards as long as it's reported upon (as Derek did). But if I can't find out because some of you seem to be against poiting such occurences out (as you are with Derek in this case), then how would the consumer know? Those are the kinds of things I expect Anandtech and other hardware sites to mention.

When it's done secretly, I don't like it, and respect Derek for beinging it up.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: ViRGE
Originally posted by: apoppin
and Virge . . . i'm sorry but your suggestion to throw away the damaged cores is silly and doesn't make any business sense for any CPU/GPU company.
I know it doesn't make any business sense. I'm just sick and tired of there being so many products that are just the same thing over and over again. The speed binning has gotten out of hand, I yearn for the simpler days where there were a couple products per core max.

the simpler days where a PC cost an average annual salary!?
:Q

yeah i remember the early 80s

i'll take the binning and cheap HW

 

acegazda

Platinum Member
May 14, 2006
2,689
1
0
They still sell the 584mhz x1900gt BTW. The 520mhz version has the x1950 pro cooler on it though. I'm prolly going to get a x1900gt considering that the x1950 pro will not be (and is not) $200 anytime soon. It took C2D 2 weeks to get below $200 for an e6300 as well.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |