prerelease benchmarks r520

Status
Not open for further replies.
Shifra said:
who fucking cares sherlock?

raise your hand if you give a shit about email times or whats in any email is fact or fiction!

now lets get one thing very clear here. There is only ONE thing someone can benefit from continously saying these are real from, ONE thing, thats getting someone to buy a GTX or GT now or making themselves feel more secure. Thats it, thats all they can gain. The cards are still not out, the NDA is still not up, these benchmarks SERVE NO PURPOSE, real or fake they are USELESS. It would be like getting benchmarks of the Voodoo 7 or the Geforce 14, who the hell cares the card is not out. By denying it you keep skeptisism where it should be, by praising it, you're being an idiot. Its that simple.
Click to expand...

Dude stop playing, you know they're real.
 
http://www.theinquirer.net/?article=26243

I know it's the Inq, but still...
ATI claims it still hasn’t sent a single R520XT board to any of its partners. The Radeon X1800XT is getting ready and the first batch of the card is produced but not even the big OEMs have these cards.

We spoke with numerous people on Friday and Saturday, and it turns out that the only cards that might be available for some key partners are Radeon X1800XL cards.

(...)

One of the key things that you must bear in mind is that R520XL, Radeon X1800XL can not be overclocked well. Voltage leakage will prevent you to increase voltage and that will limit your overclockability. This voltage leakage was the key reason for such of delay of R520. Only the flagship Radeon X1800XT will overclock like crazy. So even if some board partners tried to overclock some of the Radeon X1800 XL cards it simply would not work as you could never even touch those XT speeds of 600/1400MHz, so ATI and others claim.

There are some theories about these scores but some people at ATI who have the access to R520 hardware claim that Sanders' numbers are completely out of range. To make it even more interesting, one source confirmed that even Geforce 7800GTX are out of the range.
Click to expand...

Makes you wonder...
 
I don't wonder.

ATI has been trying to get this card to work for almost a YEAR.

It's no surprise that either the prerelease numbers are low or that they state no one has a final chip.

They've been trying to produce a "final chip" since the original "tape out" last November, what the heck do you expect? This may be the sloppiest launch in GPU history, ever.
 
Rollo said:
I don't wonder.

ATI has been trying to get this card to work for almost a YEAR.

It's no surprise that either the prerelease numbers are low or that they state no one has a final chip.

They've been trying to produce a "final chip" since the original "tape out" last November, what the heck do you expect? This may be the sloppiest launch in GPU history, ever.
Click to expand...

They have a few more months to go before they're in NV30 territory...
 
Traditionally, I don't post in this forum.

Although I have massive respect for the work that goes on with Beyond3D - I always feel that the forum is best left to enthusiasts in the general public as well as the hardcore technologists from ATI and other organisations.

However, the ‘Hardware Analysis thread’ brought up several issues that I am not about to ignore and I want to set the record straight.

Firstly, at ATI we always try to work to the highest standards of professionalism and honesty possible. We have an open relationship with all of the press across the globe and we never shy away from tough questions.

Briefings
We have invited over 100 of Europe’s top press to our R520 Tech Day which takes place at the end of September.
Even with so many places, we cannot invite everyone.
The question remains - how do we brief the press who cannot make it on the day.
From the time that our Tech Day, we will be working with the regional press that we missed to ensure that as many technical journalists as possible have access to the relevant information.
Journalists who have seen and tested the final product themselves, can then be certain of the facts, should write EXACTLY what they feel.

If they love what they see – great.

If they are not happy with what they see – also great.

All we ask for is honesty in the copy itself... that they 'write it as they see it'

Having read through enough of Dave’s reviews over the years – it is pretty clear that Beyond3D forum members have the same mind set as ATI…

JUST GIVE US THE FACTS & LEAVE THE FICTION AT THE DOOR - WE WILL MAKE OUR OWN MINDS UP THANK YOU VERY MUCH

CONCLUSION: When it comes to R520 results, you should only believe journalists who have established a reputation for being trustworthy


Not Invited
Sander has 'presented the reasons' why he was NOT on the list of 100 press for our R520 Tech Day.

However, he has conveniently left off the actual reason - which he and I have discussed at length in various emails. I am not talkning about a snatched comment - these were full blown conversations - going back and forth - that clearly detail the logic behind the decision.

Not inviting him to this Tech Day is a direct result of his response to the last invitation that I sent him

I am not going in to detail here - but it was that categorical refusal to engage that led to his being left off of this list. He knows exactly what I am talking about.

Remember, he would still have been briefed before the NDA lifted - just not at our Tech Day.

Let me say this clearly, there is no 'ATI black list for press who do not write nice stories'.

We want an open and honest press who report the facts as they see them - based on their own testing - rather than second hand gossip and fairy tales.

There are a number of reasons why any company would stop engaging with an individual - but writing honest copy is not one of them.

Honest reviews serve the vendor just as much as the customer - and anything less should be rejected.

CONCLUSION: If you choose NOT to work with a vendor - don't cry 'foul' when they decide NOT to work with you


Email conversations

As I have pointed out, honesty is crucial.

Emailing several people at ATI to explain that you are very important, that you have a series of specific demands that must be met as soon as possible and then going on to explain just how bad a job we are doing of keeping you happy is one thing.

However, including ATI's direct competition on 'cc' in such an email is another thing entirely.

To then 'jump to the moral high ground' when your 'threats' have been exposed is laughable.

CONCLUSION: Once you pick the rules of engagement - don't get upset when your true character is revealed later on


The Figures
When considering the performance of our next-generation products, you need to ask yourself one very simple question:

"Would ATI seriously bring 100 of Europe’s top press to a Technology Day where they can run whatever benchmarks they choose if we thought that we were going to lose?"

Sander claims that these numbers were given to him by a trusted source.

The numbers shown do not match any pattern that we have for our next generation products.

CONCLUSION: Either Sander lied - or his 'trusted source' set him up to look stupid on a global stage for printing made up results


Edits
Intelligent readers also need to be aware that the first version of the story he posted seemed to accuse every site who attends the ATI Tech Day of a fundamental inability to be independent in their testing and copy.

Needless to say that I am sure several sites have 'web-wacked' that original version and will be considering taking action to defend their names.

CONCLUSION: If you are going to accuse the largest independent publications in Europe of being bent - make sure you have enough money in the bank to retain good council


Logic
The whole thing has clearly been engineered - to what purpose no one can say for certain.

Maybe he is trying to force ATI to offer an invitation to Tech Day... which is not going to succeed as we refuse to be strong-armed.

Perhaps it was a simple tantrum at not being taken seriously... when, in fact, that was not the case at all.

Cynical people might say that it was to generate hits... but we cannot be sure of that either.

Whether it was one of these reasons - or a completely different twist of logic - one thing is for certain - it has certainly acted as an appetiser for ATI's next generation launch... and for that I have to say 'Thanks Sander !'

Having corrected several of the 'inaccuracies', I am not going to say any more on the subject because the real strength of the Beyond3D forums is their focus on technology.

How open are we going to be with our Tech Day ?

Simple, in addition to the top news & reviews sites, we have also invited one of the largest nVidia fan sites in the world to come along, get briefed and benchmark R520 however they choose - along with the best of them.

The world's top press will have hands-on our latest products in less than two weeks and you can all decide then if you prefer facts or fiction.

Thank you for taking the time to read this

Kind regards,

Andrzej Bania
ATI Technologies
Click to expand...

Taken from B3D.

Who do you believe? I know who I do.
 
such a mess lol

im so glad i decided to never buy an ati card, if its not opengl issues, its drama gah.

move along kids nothing to see here
 
fallguy said:
Taken from B3D.

Who do you believe? I know who I do.
Click to expand...

Thanks for the link, fallguy. Hadn't seen this. Even if it states the obvious - you should plaster all vid-card forums on the net with this to get things back into proportion again :D

has there ever been a worse rumor-war before a product launch? It's always been harsh, but this time...
 
Wow, even a big Nvidia fansite is invited to Ati's R520 tech day. I don't think Ati is doing this to show them something to laugh about. :)
Looks like Ati is very confident. :cool:
 
or hes trying to save face, and his job in front of the board of directors who im sure are closely watching all of this
 
emailthatguy said:
or hes trying to save face, and his job in front of the board of directors who im sure are closely watching all of this
Click to expand...

So you don't believe him? Wow, can they ever tell the truth, or does there have to be another secret motive in every statement they make?
 
Bottom line? It's his word against the other guy's. "E-mails" like that could be manufactured. This could all be lies.

We'll know the truth in a couple weeks I guess.
 
fallguy said:
Taken from B3D.

Who do you believe? I know who I do.
Click to expand...

Fallguy believes the ATI PR rep over a reviewer posting something negative about ATI?!?!?

:eek:

Impossible!

;)
 
Rollo said:
Fallguy believes the ATI PR rep over a reviewer posting something negative about ATI?!?!?

:eek:

Impossible!

;)
Click to expand...


the GTX numbers are messed up too rollo....

i cant help but to chuckle at some people. After all this you turn into tin foil hat conspiracy, i wonder if the earth is flat and the moon is cheese in your world too.
 
Riptide_NVN said:
Bottom line? It's his word against the other guy's. "E-mails" like that could be manufactured. This could all be lies.
.
Click to expand...

Uhh, even SS didn't deny the e-mails, he said that he was making more or less "a joke" with that "lowering stock" thingy.
Wow what a joke it is.
 
Rollo said:
Fallguy believes the ATI PR rep over a reviewer posting something negative about ATI?!?!?

:eek:

Impossible!

;)
Click to expand...

And Im not alone. The VAST majority dont believe Sanders. Not just normal forum goers like us, but other hardware reviewers, and smart minds in the buis. Especially since Andrzej offered he would be, "definitely be up for discussing the whole thing out in the open", with Sanders. He is willing to lay his cards on the table, backed up with emails between the two. Sanders hasnt accepted.

So tell me. Who do you believe, Saders, or the guy from ATi? Simple question, dont dodge it.
 
fallguy said:
And Im not alone. The VAST majority dont believe Sanders. Not just normal forum goers like us, but other hardware reviewers, and smart minds in the buis. Especially since Andrzej offered he would be, "definitely be up for discussing the whole thing out in the open", with Sanders. He is willing to lay his cards on the table, backed up with emails between the two. Sanders hasnt accepted.

So tell me. Who do you believe, Saders, or the guy from ATi? Simple question, dont dodge it.
Click to expand...

I'll take those benchmarks as valid until I see contradictory ones from a site I trust.
 
fallguy said:
And Im not alone. The VAST majority dont believe Sanders. Not just normal forum goers like us, but other hardware reviewers, and smart minds in the buis. Especially since Andrzej offered he would be, "definitely be up for discussing the whole thing out in the open", with Sanders. He is willing to lay his cards on the table, backed up with emails between the two. Sanders hasnt accepted.

So tell me. Who do you believe, Saders, or the guy from ATi? Simple question, dont dodge it.
Click to expand...

Tell us, what you say IF those benches turn out to be right on or very close? I will publicly apologize to everyone. Would you admit that you were wrong? I will admit since they are not official, the safe bet would be to not look upon them as valid. From my perspective though, see how the R520 is built, I believe it.

I also believe the R520 is part R520 and part R420. It doesn't matter than ATi said it was an "all new" architecture. Guess what nVidia said the same thing about the G70, but guess what, the G70 is a modified NV40! Sure it's more efficient and faster than the 6800 but it's made from the 6800 and why not. I just don't like it when anyone lies, be it ATi or nVidia. Also, look at this, the G70 a 2nd gen card (to the NV40) on a 110nm process has 24 pipes and has allot of headroom to O/C. The R520 built on a 90nm process only has 16 pipes and only the XT can O/C well. The XL as well all the others don't O/C well at all. http://www.theinquirer.net/?article=26243 If the R520 was built on a all new architecture, it would have any problem O/Cing or having 24 or 32 pipes. Why do you think the R580 has 24 and 32 pipes...it actually is all new!

That's the reason ATi had such a hard time with so many tape outs! I think originally they had planned the R520 to be the R580 but when they saw it would take more time to develop they went to plan B which was the R520. We'll just built a hybrid and that should be a good stopgap until we can finish the R580. I know....I know....the R580 has already been taped out right...well ATi didn't know they would finish so quickly. Anyway I think my theory is correct and if those scores are right or close to it, that will support it even more! It's either the hybrid theory or ATi just plain sucks! I think the R520 is plan B and it's a hybrid. Watch, if it is, it will come out!
 
Rollo said:
I'll take those benchmarks as valid until I see contradictory ones from a site I trust.
Click to expand...

Wow, I didnt expect that from you.

Bowen said:
Tell us, what you say IF those benches turn out to be right on or very close? I will publicly apologize to everyone. Would you admit that you were wrong?
Click to expand...

Wrong about what? I never said the benches were real, or fake. I said several times that there is no proof for us general public either way. I dont believe anything thats on the internet. Apparently many others do.

What I did say is that I believe Andrzej over Sanders, about what happened leading up to, and with the "review".
 
fallguy said:
Wow, I didnt expect that from you.



Wrong about what? I never said the benches were real, or fake. I said several times that there is no proof for us general public either way. I dont believe anything thats on the internet. Apparently many others do.

What I did say is that I believe Andrzej over Sanders, about what happened leading up to, and with the "review".
Click to expand...

I don't believe Andrzej at all, if anything both these guys are equally mud slingling neophants.

Until Sanders provides benchmark scripts his validity is seriously under question too. But Andrzej seems to be going very far with his accusations which really are borderline liable. No let me rephrase that it is liable.
 
Terra said:
The plot thickens *LOL*

Terra - Can't wait to see [H]ardOCP's benches...
Click to expand...

Holy Moly! Have never seen anything like this in all the years that I have been following hardware. Nvidia must be loving it
 
defiant said:
Holy Moly! Have never seen anything like this in all the years that I have been following hardware.
Click to expand...

no shit! in-fucking-credible... soap-opera


defiant said:
Nvidia must be loving it
Click to expand...


only if r520 does not perform better than their silicon and this was not just the ridiculous and childish revenge of some poor kid who didn't get invited along to a party.
 
Rollo said:
I'll take those benchmarks as valid until I see contradictory ones from a site I trust.
Click to expand...

Well, that's just ridiculous. He didn't run the benches himself, he cannot even confirm their veracity.

Even Kyle has said Sanders needs to own up.

There is no logic to your position that I can see.
 
honestly its not like crossfire is going to change the world

i mean comon 2 sku's for video cards? maybe like 4-5 major retailers will get behind this but nowhere near large volume. its a dumb idea really. availability is going to suck because the market is so niche its not even funny. dell wont be buying into this, neither will gateway or any other major oem which is really what drives the market. hell id be shocked if alienware or voodoopc got behind it.

in reality performance numbers aside. this is going to be a product that maybe at best 2% of the market gets behind if theyre lucky.

in 2 years theyll discontinue it if not before due to insanely low sales since video cards for it are going to be tough to find. and we'll all have forgotten about this madness.
 
emailthatguy said:
honestly its not like crossfire is going to change the world
Click to expand...
The more of Xfire that I see the more I think ATI should have just sucked it up and adopted NV's SLI instead.
 
Jonsey said:
Well, that's just ridiculous. He didn't run the benches himself, he cannot even confirm their veracity.

Even Kyle has said Sanders needs to own up.

There is no logic to your position that I can see.
Click to expand...

If you have a friend at an OEM who says he ran benches for you, why wouldn't you believe him?

The HA benches may be fake, they may be real. No one can say for sure until other people prove or disprove his findings.

Kyle may think Sassen needs to "own up" but unless Kyle is saying that from the perspective of someone who has benched those cards, he's in the same position you and I are: We don't know yet.
 
Rollo said:
If you have a friend at an OEM who says he ran benches for you, why wouldn't you believe him?

The HA benches may be fake, they may be real. No one can say for sure until other people prove or disprove his findings.

Kyle may think Sassen needs to "own up" but unless Kyle is saying that from the perspective of someone who has benched those cards, he's in the same position you and I are: We don't know yet.
Click to expand...

I think I have this figured out...

The benchmarks Sanders got were done at the "XT clocks on pre-production hardware"...so maybe the reason the numbers don't line up to what ATi says is because they haven't finalized the clocks yet...and they are pushing them higher to compete with the GTX...

This would make sense give the inquirer article today about the clock speeds being higher than 600Mhz...and also line up with what ATi's CEO said about how they should be able to beat the gtx if they get the right clocks...

So in the end...nobody is lying...
 
razor1 said:
I don't believe Andrzej at all, if anything both these guys are equally mud slingling neophants.

Until Sanders provides benchmark scripts his validity is seriously under question too. But Andrzej seems to be going very far with his accusations which really are borderline liable. No let me rephrase that it is liable.
Click to expand...
You mean libel.
And no, nothing I've seen Andrzej say can be remotely construed as libel.
It's possible what Sanders has done, though, if he is found to be lying, is. Falsifying bencmarks to hurt ATi stock, could be construed as libel.
 
I originally posted this response on my website (link) after writing up an article on the ongoing debacle. This is what I had to say about today's events:

Allow me to start off by saying that it's hard not to agree with Kyle on such pressing matter. He is, of course, an icon in the eyes of gamers and hardware enthusiasts far beyond the internet. Kyle has been praised time and time again for his "take no shit" attitude and isn't someone that you or I would want to cross in a courtroom.

The issue isn't whether Kyle's response is valid, and I want to make it clear that the point of drawing out this write-up was to get an idea of how our readers feel about benchmarking and hardware reviews as whole. I feel like we, as consumers, fail to look below the surface of a given review just to fall victim to corporate propaganda. When we see those bar charts lined up neatly in the center of a webpage, our eyes are automatically drawn to them not because of their colors, but rather because of both our curiosity and desire for an answer.

I'll admit that I've been a victim too, and to this day I remain extremely skeptical about the reviews I see popping up on the internet for not only hardware, but also for games. At one time, File Rush News (AIXGaming) used to display the latest reviews from sites like Yahoo!'s DomainOfGames.com and occasionally GameSpy. Now, that's not to say that my trust in either of those sites has completely dissipated away to nothing over the years, but at the same time… I've lost *some* faith in online journalism.

So who should you trust and how far should you trust them? That's entirely up to you, the reader. I try to be extremely careful about what I report on simply because I don't want anyone buying something or trying something out that isn't up to par with the comments that I've awarded the product being used. When I say not to buy a Gizmondo, for example, I'm not saying it to spite Tiger and destroy what little marketshare they possess, but what I am saying is that our readers face being "screwed over" if they don't wait for the Widescreen model in 2006.

With that said, I leave the floor open for debate. If you don't have answers to the questions I've imposed, then feel free to talk about whatever comes to mind.
Click to expand...
Obviously the context of my statement fits my post a bit better than this thread, but hopefully everyone sees where I was trying to go with it. Online journalism has always been a tough cookie to crumble, and it's obvious that a *lot* of people aren't cut out for the job. When it comes down to weeding them out however, it could take something very little like a barrage of bad feedback or something catastrophic like posting a fake benchmark / posting an inflated review of a game on your website.

In other words, the purpose of bringing up Sander's actions should bring to question the validity of the work being performed on other sites as well. I'm not saying that you should go around and interrogate respected members of the gaming/hardware communities for posting outrageous material, but what you should do is keep your guard up and be on the lookout for rubbish.
 
Ateam311 said:
I originally posted this response on my website (link) after writing up an article on the ongoing debacle. This is what I had to say about today's events:


Obviously the context of my statement fits my post a bit better than this thread, but hopefully everyone sees where I was trying to go with it. Online journalism has always been a tough cookie to crumble, and it's obvious that a *lot* of people aren't cut out for the job. When it comes down to weeding them out however, it could take something very little like a barrage of bad feedback or something catastrophic like posting a fake benchmark / posting an inflated review of a game on your website.

In other words, the purpose of bringing up Sander's actions should bring to question the validity of the work being performed on other sites as well. I'm not saying that you should go around and interrogate respected members of the gaming/hardware communities for posting outrageous material, but what you should do is keep your guard up and be on the lookout for rubbish.
Click to expand...

Well said...greets from charlottesville...:)
 
Terra said:
The plot thickens *LOL*

Terra - Can't wait to see [H]ardOCP's benches...
Click to expand...

sanders said:
In the end, for reasons undisclosed, ATI however never followed through with this event.
...
Click to expand...

Very interesting. Wonder why they canceled that event. After all this...wouldn't suprise me a bit if there is an X1800XTPE or the analog...at 650 core. Good luck finding one, though. What everyone needs to remember at the end, is whether the sanders bench scores line up with the 16 pipe 1800 at 600mhz core clock, using the drivers available at the time
 
^eMpTy^ said:
Well said...greets from charlottesville...:)
Click to expand...
Howdy. I'm actually down in Blacksburg @ VT for the year, but I live in H'burg for the remainder of the year.

Glad you agree with what I said btw. :)
 
eno-on said:
You mean libel.
And no, nothing I've seen Andrzej say can be remotely construed as libel.
It's possible what Sanders has done, though, if he is found to be lying, is. Falsifying bencmarks to hurt ATi stock, could be construed as libel.
Click to expand...


Wrong read the B3D thread again.
 
^eMpTy^ said:
I think I have this figured out...

The benchmarks Sanders got were done at the "XT clocks on pre-production hardware"...so maybe the reason the numbers don't line up to what ATi says is because they haven't finalized the clocks yet...and they are pushing them higher to compete with the GTX...

This would make sense give the inquirer article today about the clock speeds being higher than 600Mhz...and also line up with what ATi's CEO said about how they should be able to beat the gtx if they get the right clocks...

So in the end...nobody is lying...
Click to expand...

Well, many have said the benches on HA don't match the pattern of an R520 card, that is, they don't correctly show the areas the R520 would be strong and weak.
 
Jonsey said:
Well, many have said the benches on HA don't match the pattern of an R520 card, that is, they don't correctly show the areas the R520 would be strong and weak.
Click to expand...
I thought the card was still MIA so how can we actually know where the card is "strong and weak" when it's suppose to be an architectural change from the X8x0 series? ;)
 
CrimandEvil said:
I thought the card was still MIA so how can we actually know where the card is "strong and weak" when it's suppose to be an architectural change from the X8x0 series? ;)
Click to expand...

Cause ATI said so. Or that's what's been told to us by ATI I mean.
 
razor1 said:
Wrong read the B3D thread again.
Click to expand...
Um, no. I'm not wrong. If what you said isn't what you mean, then learn how to say what you mean.
 
Status
Not open for further replies.
Back
Top