More HL2 Propaganda from Anand

[WaltC]

http://www.anandtech.com/weblog/index.html?FTBLOGDATE=10/01/2003

Here are some Half Life 2 numbers for you to look at; they were provided by a reliable source, but I could not verify anything myself so take them with a grain of salt. ATI was running in their DX9 codepath and the mixed mode codepath was used for NVIDIA. No AA/AF was enabled and we're looking at 1024x768 scores:

(See link to get scores.)

If those numbers hold true then things definitely look better than from Half Life 2 day, but we'll reserve judgement until we get the benchmark in house. I just thought you'd like to see what we're seeing, I wouldn't draw any conclusions based on this data yet, just wanted to share

So--now Anand is printing speculation based information he admits he cannot verify in any way, shape, or form. What can this possibly represent except that Anand is attempting to spin up nVidia, and plug them? If he really wanted to "reserve judgement" he would have done so and waited until he got the benchmark in house, of course. This is nothing but propaganda, apparently. It's nice that he admits the two products weren't running the same code paths, but that's about all I can say about it. Why on earth would you run numbers you cannot verify?

OK, now I'm hearing rumors that the HL2 source code has been stolen and distributed on the 'net. Either nVidia has a mole in Valve/Vivendi, or...I don't know what.

WHAT'S HAPPENED TO INTEGRITY? DOESN'T ANYONE HAVE IT ANYMORE?

(If this has been posted before, apologies. I checked for a similar thread and only found one based on another Anand tidbit he printed earlier.)

 

[Joe DeFuria]

http://www.beyond3d.com/forum/viewtopic.php?t=8221&start=20

(Recently brought up in the bottom of the thread.)

Why on earth would you run numbers you cannot verify?

Because nVidia gave Anand special treatment wrt to NV38 NDA, so Anand gives nVidia a little special treatment in return.

 

[jjayb]

There is no reason on earth he should be posting those numbers. He didn't run them. I wonder who is "reliable source" is. Could it be.....oh, I don't know............Nvidia!?! Would they have told him if the graphics were borked?

 

[CorwinB]

I could not verify anything myself so take them with a grain of salt.

I suppose Anand is expecting us to take those scores with a grain of salt, and a hook, a line and a sinker too ?

I just thought you'd like to see what we're seeing

Last week, he was hearing. This week, he's seeing. If things keep getting better, then maybe he will actually start testing within a few years.

 

[Bouncing Zabaglione Bros.]

There is no reason on earth he should be posting those numbers. He didn't run them. I wonder who is "reliable source" is. Could it be.....oh, I don't know............Nvidia!?! Would they have told him if the graphics were borked?

This reminds me of when the first NV30 benches sneaked out, and Nvidia was adamant that the tests were only to be taken at a particular res with no AA/AF. Unsurprisingly, they made the NV30 look much, much better than it was.

Here we go again with unsubstantiated leaks (supplied via a well know Nvidia mouthpiece), with numbers way, *way* above what we are expecting to see, and we are supposed to believe this crap? When the game and the 5950 arrives, we'll see that Nvidia IQ is severely degraded, that no one reaches those numbers, and they'll drop by three-quarters anyway when enabling AA/AF.

 

[RussSchultz]

OK, now I'm hearing rumors that the HL2 source code has been stolen and distributed on the 'net. Either nVidia has a mole in Valve/Vivendi, or...I don't know what.

Those are some strong and unfounded accusations. Will you blame the next earthquake on NVIDIA, also?

 

[FUDie]

Here we go again with unsubstantiated leaks (supplied via a well know Nvidia mouthpiece), with numbers way, *way* above what we are expecting to see, and we are supposed to believe this crap? When the game and the 5950 arrives, we'll see that Nvidia IQ is severely degraded, that no one reaches those numbers, and they'll drop by three-quarters anyway when enabling AA/AF.

These numbers may be legit, but notice that the NV38 is running the mixed-mode path. The whole point of that path was to raise the performance of the NVIDIA parts so that they would get playable framerates. Also, given that the results are nearly identical in many cases, with different code paths on different hardware, it seems likely that at least some of the results are CPU limited.

"take these with a grain of salt": Why post them then? If you know the results are not verifiable, then you shouldn't post it. What if someone else sends Anand a table of results showing R360 to be 2 or 3 times faster than the NV38 on the DX9 path? Would he post those results too?

This is so ridiculous...

-FUDie

 

[Borsti]

http://www.beyond3d.com/forum/viewtopic.php?t=8221&start=20

(Recently brought up in the bottom of the thread.)

Why on earth would you run numbers you cannot verify?

Because nVidia gave Anand special treatment wrt to NV38 NDA, so Anand gives nVidia a little special treatment in return.

Well, Anand was not the only site who got this offer. But he is (as far as I saw) the only site who did the NV38 testing. We did´nt like the idea for several reasons... so we did´nt include any NV38 numbers and Det 52 in the article.

Lars

 

[WaltC]

OK, now I'm hearing rumors that the HL2 source code has been stolen and distributed on the 'net. Either nVidia has a mole in Valve/Vivendi, or...I don't know what.

Those are some strong and unfounded accusations. Will you blame the next earthquake on NVIDIA, also?

They are not accusations, Russ, they are reasonable suppositions. First of all, we don't know if the HL2 source has actually been stolen--so that needs to be determined first. No, I don't think nVidia will have anything to do with the next earthquake, so that should put your mind at ease (since you have an obvious sensitivity, apparently, to what anyone might possibly be thinking about nVidia at any time...)

I think it's very reasonable to pose the question as to who might benefit from the benchmark scores Anand regards highly enough to print while admitting at the same time he hasn't a clue as to their validity. There's nothing "newsworthy" about that at all. At best, it's hucksterism, at worst it's fraud. Also, he claims the information comes from a "reliable source" which he finds unreliable enough to refrain from naming.

So, OK, who would have the "credibility" to influence Shimpi to print this stuff, and at the same time hope to see a benefit from this very narrow little slice of a "benchmark" result getting posted publicly? Let's see, there are only two IHV products featured in the "benchmark." I think we can dismiss ATi as having any motivation whatsoever to pose as Shimpi's "reliable source."

Next, what benefit does it do Shimpi's credibility to "share" benchmark scores with his readers while he admits he hasn't the foggiest as to whether they are even legitimate? No system specs--nothing to provide any meaningful info there at all. Why bother to state that your source is "reliable" when the information he provides is incomplete to the point of being meaningless? So, why go out on a limb, with text published under your own name, to publish "scores" you *admit* you cannot verify? In other words--Shimpi is publishing "benchmark" data on his own site that he admits could be worthless. So if you do not know whether it is worthless, why publish it? Is it professional of him to publish something that *might be* true--gee, if only he knew? Well, I guess it's barely possible Shimpi is an amiable dolt, thick between the ears, who is being used and is too stupid to know it? You think that's likely? I don't.

No, what I think is that Shimpi knows exactly who the source is, exactly what the source has to gain from something like this, and is consciously complicit with that source in the disemination of "information"--that out of his own mouth--Shimpi *will not* swear is genuine. Again--if you don't know it's genuine--why publish, especially under your own name?

Why do you think he published it, Russ?

 

[Joe DeFuria]

Well, Anand was not the only site who got this offer. But he is (as far as I saw) the only site who did the NV38 testing. We did´nt like the idea for several reasons... so we did´nt include any NV38 numbers and Det 52 in the article.

Lars

Can you clarify?

What specific offer are you saying you were handed, yet refused to publish: these same Half-Life2 benchmarks, or an offer to benchmark NV38 / Det 52?

 

[Borsti]

The NV38/Det52 benchmarking. No idea about the HL2 benchmark numbers.

Lars

 

[Joe DeFuria]

The NV38/Det52 benchmarking. No idea about the HL2 benchmark numbers.

Lars

Well, then that exactly agrees with my suspicions.

1) Anand took nVidia up on it's offer to "pre-benchmark" NV38 to be included in a Radeon 9800 XT review...you didn't.

2) Anand comes up with Half-Life2 benchmarks from "a reliable source"....you didn't.

BTW...kudos to you for refusing to run such benchmarks upon nvidia's offer. I would have thought that running benchmarks on unannounced hardware, using non-released drivers would be a no-brainer *cough* Anand *cough*.

 

[jimbob0i0]

Well HL2 source WAS stolen... Gabe's confirmed it at halflife2.net ..... was a combination of Outlook exploit, keylogger and remote control client... poor guy

 

[WaltC]

The NV38/Det52 benchmarking. No idea about the HL2 benchmark numbers.

Lars

Well, then that exactly agrees with my suspicions.

1) Anand took nVidia up on it's offer to "pre-benchmark" NV38 to be included in a Radeon 9800 XT review...you didn't.

2) Anand comes up with Half-Life2 benchmarks from "a reliable source"....you didn't.

BTW...kudos to you for refusing to run such benchmarks upon nvidia's offer. I would have thought that running benchmarks on unannounced hardware, using non-released drivers would be a no-brainer *cough* Anand *cough*.

Well, from what I've seen in the last few minutes all over the 'net--it's true, Valve was hacked and the source code was stolen from right under their noses. Here's Gabe Newell's text as reprinted on BLue's News:

Ever have one of those weeks? This has just not been the best couple of days for me or for Valve.

Yes, the source code that has been posted is the HL-2 source code.

Here is what we know:

1) Starting around 9/11 of this year, someone other than me was accessing my email account. This has been determined by looking at traffic on our email server versus my travel schedule.

2) Shortly afterwards my machine started acting weird (right-clicking on executables would crash explorer). I was unable to find a virus or trojan on my machine, I reformatted my hard drive, and reinstalled.

3) For the next week, there appears to have been suspicious activity on my webmail account.

4) Around 9/19 someone made a copy of the HL-2 source tree.

5) At some point, keystroke recorders got installed on several machines at Valve. Our speculation is that these were done via a buffer overflow in Outlook's preview pane. This recorder is apparently a customized version of RemoteAnywhere created to infect Valve (at least it hasn't been seen anywhere else, and isn't detected by normal virus scanning tools).

6) Periodically for the last year we've been the subject of a variety of denial of service attacks targetted at our webservers and at Steam. We don't know if these are related or independent.

Well, this sucks.

What I'd appreciate is the assistance of the community in tracking this down. I have a special email address for people to send information to, helpvalve@valvesoftware.com. If you have information about the denial of service attacks or the infiltration of our network, please send the details. There are some pretty obvious places to start with the posts and records in IRC, so if you can point us in the right direction, that would be great.

We at Valve have always thought of ourselves as being part of a community, and I can't imagine a better group of people to help us take care of these problems than this community.

Gabe

 

[Joe DeFuria]

It's interesting to note that the first apparent signs of trouble (9/11) was just after both of the following events occurred:

1) Valve promotes ATI hardware with benchmarks
2) Steam exits beta stage and goes live.

 

[Borsti]

I had my reasons. Your "btw" comes rather close to it

What I meant regarding the HL2 numbers. I did not receive those results. And if they came from NV (what I don´t know) I would see no reason why they would not like to be named as source. NV has an interrest to let the people know that they´re working on HL2 perfomance. So maybe they came from another source!?

Lars

 

[Sabastian]

This may seem a stupid question to ask but who would be interested in having Valves HL2 source code? Whom would benefit most from having it?

 

[Joe DeFuria]

I had my reasons. Your "btw" comes rather close to it

Good to hear!

I would see no reason why they would not like to be named as source.

Other than the obvious: Valve probably forbids ANYONE running and publishing bencharks at this time. Valve has certainly forbid anyone from testing nVidia cards with anything other than the officially available drivers. So certainly, Valve would be even MORE PISSED at nVidia if it was found out that nVidia passed the numbers (using unapproved drivers) to anyone for publication.

So maybe they came from another source!?

Always a possibility...but why?

 

[CorwinB]

Well, Anand was not the only site who got this offer. But he is (as far as I saw) the only site who did the NV38 testing. We did´nt like the idea for several reasons... so we did´nt include any NV38 numbers and Det 52 in the article.

Lars

The times they are a-changin' quickly it seems...
Just to think that not so long ago THG was "happy to serve" with the Doom3 benchmarketing charade. And even more recently, the Good Dr. himself was kind enough to enlighten us with how a poor IHV is victim from an atrocious conspiracy. Of course, since we are all morons brainwashed by the ATI PR guys, the Good Dr. absolutely had to resort to the lowest possible analogies (including references to Al Quaida and Saddam Hussein) in order to rise an healthy indignation in our heart. Thank you so much, Good Dr.
[/url]

 

[Bouncing Zabaglione Bros.]

I had my reasons. Your "btw" comes rather close to it

What I meant regarding the HL2 numbers. I did not receive those results. And if they came from NV (what I don´t know) I would see no reason why they would not like to be named as source. NV has an interrest to let the people know that they´re working on HL2 perfomance. So maybe they came from another source!?

Lars

It's an Nvidia standard operating procedure. When they want to run spoilers or leak out information, they do it silently via a third party so that they can deny anything that has been said or published.

Look at what happened when Nvidia "promised" Kyle at [H] that trilinear would be restored. Kyle told everyone else it would all be fixed in the next driver and was a complete non-issue, and then several weeks later, Nvidia can simply ignore the commitment was publicly made in their name, and reduce IQ even further whilst leaving everyone with the impression it was a bug that was being fixed.
Look at the "bench with Det 50's for speed, but don't talk about them or issue them to customers".

Nvidia love having their "plausable deniability".