NVIDIA claims top DX9 GPU marketshare spot.

[Dave H]

demalion-

I understood marketshare numbers as the context for Jerky's "properly" remark because they were the basis for the thread (which he started) and given in its title (which should be considered also as the title for that post, the first in the thread). In particular, something about the word "claims" in the thread title made it seem like Jerky was questioning the factual basis for the claim. I understand from your reply that you would agree a "claim" of DX9 marketshare should indeed include 5200 sales.

Having said that, looking back at Jerky's post I may have misinterpreted him. He does not specifically say that the marketshare numbers should not include 5200 sales, but only that by including them they are "misleading". I took "misleading" to be a normative charge that the numbers are essentially wrong, and that the basis for calculating them should be changed. I disagree with that assertion, and I gather that you would too.

However, "misleading" could also be taken to mean simply that the numbers don't tell the whole story. I would generally agree with that sentiment, and I'm pretty sure you would also. If Jerky meant "misleading" in that sense, then my criticism of his post was unnecessary and I apologize for the misunderstanding.

As for the rest of your post, I'm not going to respond to it, not because it's not worth discussing, but because I simply don't have the time (and likely won't, at least not for several days).

 

[RussSchultz]

What happened to the end of the obligatory put downs?

Look, I stated my opinion, and clarified it for you with my reasoning.

There's no reason to spend your time or mine writing paragraph after paragraph about how I cannot comprehend the english language and how "wrong" I am in my opinion.

And I'm completely serious, and not trying to insult you with the critique on your conversational style. You respond with WAY to much verbeage, and it comes across as trying to bury the other person's point/opinion with volume.

If you feel what I originally had to say was "merely a potshot", then say that, and not an essay on how its wrong, describing in many details how its wrong, and somewhere in there saying "oh, and its a potshot too".

It would save me, and you a lot of time if you simply got to the point.

Beyond that, since we're discussing an opinion here (mine), I don't really see a need to answer your essay point by point. I stated why I believe such an opinion, and you've stated how you think its wrong. After two times around the mulberry bush, just assume I disagree with your assessment of "wrongness".

You're not going to change my mind. Simply because I don't want to go back and forth restating my opinion and having you try to change it by being presented with your reasoning along with intellectual slights doesn't mean I'm stupid or refuse to acknowledge that I'm wrong. I just simply disagree with your differing opinion.


Completely off topic: I've noticed that "strawman" is the coup de grace de jour, as it were. Everything is struck down as a strawman. I wonder how long until "strawman" becomes as unacceptable as "fanboi".

 

[demalion]

...

Yes.

...

Yes.

...

Yes.

However, "misleading" could also be taken to mean simply that the numbers don't tell the whole story. I would generally agree with that sentiment, and I'm pretty sure you would also. If Jerky meant "misleading" in that sense, then my criticism of his post was unnecessary and I apologize for the misunderstanding.

Please note your initial post and the followup were far away from establishing this issue, and I'd propose that their presented reasoning still would not stand even if the criticism of Jerky's viewpoint did. A quote of the use of "misleading" would, however (IMO).

As for the rest of your post, I'm not going to respond to it, not because it's not worth discussing, but because I simply don't have the time (and likely won't, at least not for several days).

OK.

 

[demalion]

What happened to the end of the obligatory put downs?

A put down is a taunting or posturing statement made to describe someone unflatteringly and provide a platform for argument-by-ridicule. The first sentences are the only instance of it I'm aware of, by way of what seems your preferred post "greeting". The rest of my post consists of pretty accurate descriptions of your actions and questions...do you have a specific example in mind?

Look, I stated my opinion, and clarified it for you with my reasoning.

No, you didn't clarify it for me with reasoning. Your reasoning was flawed...can you process that statement and how it works against your "clarifying" something?

...Russ says my posts are too long again...

If you feel what I originally had to say was "merely a potshot", then say that, and not an essay on how its wrong, describing in many details how its wrong, and somewhere in there saying "oh, and its a potshot too".

Actually, that's all I did say to you! The rest of the initial post was directed at Dave H, and it was pretty short.
(talking about what I was criticizing about what Dave H seemed to propose):"...Russ was correct to propose that this opinion indicated incompetence (which, I'll mention since I'm responding here, did not seem to be an at all useful way of proposing disagreement itself )".

Do you have a point?

The expanded reply was because of the extreme lack of sense you made in your reply to my statements. A reminder: you said that simply because Jerky mentioned performance "specifically", that he obviously couldn't be including the criteria of performance in "properly". Some sort of one word jeopardy that seems to have been made up on the spot, AFAICS. Details discussed prior.

It would save me, and you a lot of time if you simply got to the point.

Russ, I get to the point...for each new point you tack on (like this one about me not getting to the point). It would save time if you responded to my points the first time, and with a useful point of your own, instead of throwing another assertion in front of it that I have to dig through to get back to being allowed to address your commentary again.

All of those sentences had a point, Russ. There are a lot of problems in what you are saying.

Beyond that, since we're discussing an opinion here (mine), I don't really see a need to answer your essay point by point. I stated why I believe such an opinion, and you've stated how you think its wrong. After two times around the mulberry bush, just assume I disagree with your assessment of "wrongness".

The issue is that your criteria for disagreement does not include things that seem to be true, or demonstrate an interest in trying to include such things.
I propose that behaving in such a way is going to earn you criticism, and that perhaps you deserve it. Am I wrong?

You're not going to change my mind.

You maintain that you're free not to change your mind about what you say, but you also maintain that I must change mine about what I say? To defend such an assertion other than by ego (i.e., the world should suite you over anyone else), you have to explain why and show that the why makes sense. One of us in this conversation is doing this, IMO, and the other is complaining about it. I'm open to have my mind changed, but just maybe the facts indicate I shouldn't. If you disagree, please give some illustration...otherwise you're telling me the world is flat.

But perhaps we should discuss the flatness of the world in some other thread. Feel free to return to the discussion about your proposing that Jerky is incompetent at some point in this thread. Hey, maybe even apologize if you might have been unfair?

Simply because I don't want to go back and forth restating my opinion and having you try to change it...

Change it? I'm pointing out how it is in error. If you can be shown how it is in error, and don't care to modify your opinion, I think that's a problem in you. Hey, but it's not your opinion, I can't say it at all, even if I provide reasons.

...by being presented with your reasoning along with intellectual slights doesn't mean I'm stupid or refuse to acknowledge that I'm wrong.

I didn't say you were stupid, nor in my experience does your behavior require stupidity. Maybe our conversations would be briefer if you stopped adding new things for me to correct.
As for refusing to acknowledge that you're wrong...

I just simply disagree with your differing opinion.

What about the things that I proposed as facts? You seem to have just avoided them repeatedly.
So...all opinions are created equal, purely by virtue of being held by someone? I'd say that the opportunity to express an opinion is equal instead...that's not the same thing.

 

[nelg]

A waiver is effectively an area of non-compliance but MS are willing to overlook for certain reasons. I'm hearing that the 5200 has several waivers - we know that it only runs in FP16, so this is likely to be one of them.

The WHQL process is quite convoluted, not necessarily in terms of testing for compliance but more in terms of penalties for non-complaince (i.e. for non-WHQL hardware saying windows compatible being sold, the vendor has to pay MS money for each unit).

Dave, does MS make waiver information public or at least a list of what areas inwhich non-compliance is accepted under waiver ?

 

[Evildeus]

That doesn't answer my qestion. Is it or not qualified by MS to be called DX9 card (ie compliant), and are the drivers WHQL?

Well, it does answer your question since MS have stated what the specification is, and it evidently hasn't met it.

No it doesn't. That's what you say, and it's perfectly understable. But what i want is the specific statement from MS saying it is or not DX9 compliant.

 

[Dave Baumann]

Well, you've seen the statement from MS that FP24 is the minimum requirement for the Pixel Shader haven't you? That is the answer.

As for the difference between market share and accountancy: NVIDIA only recognises revenue on units shipped by the AIB (i.e. actual units sold), Mercury research numbers account for units shipped by the IHV, not units sold.

 

[Evildeus]

Well, you've seen the statement from MS that FP24 is the minimum requirement for the Pixel Shader haven't you? That is the answer.

That not an answer. It could (and should) disqualify the FX5200, but it finally depends on MS decision which may forget this part. That's why i'm looking for a definitive statement from MS. Not some partial answers.

As for the difference between market share and accountancy: NVIDIA only recognises revenue on units shipped by the AIB (i.e. actual units sold), Mercury research numbers account for units shipped by the IHV, not units sold.

Of course, that's why unit sold is always > as unit shipped (but at the end of the life it's equivalent). I don't understand what you are trying to say?

 

[Hanners]

That not an answer. It could (and should) disqualify the FX5200, but it finally depends on MS decision which may forget this part. That's why i'm looking for a definitive statement from MS. Not some partial answers.

I think you're going to have a very long wait then. This isn't exactly the thing that MS are going to comment on is it? (Especially if a waiver was involved). I think on this occasion you just need to deduce the conclusions from the information you have been given here.

 

[Dave Baumann]

That not an answer. It could (and should) disqualify the FX5200, but it finally depends on MS decision which may forget this part. That's why i'm looking for a definitive statement from MS. Not some partial answers.

The statement is there - the specifications are there, testing shows that they are not to those specificatrions. Stop being obtuse.

Of course, that's why unit sold is always > as unit shipped (but at the end of the life it's equivalent). I don't understand what you are trying to say?

Units sold can never be greater than units shipped! I assume you mean "unit sold is always < unit shipped". However, it's harking back to you statement that NVIDIA recognises a unit sale ones its sold by the vendor - it does not apply here.

 

[Evildeus]

Well sorry to be obtuse, but that statement from MS doesn't automatically invalidate the possibility that the 5200 is a DX9 compliant card for MS. Anyway, where can i find the link where it shows that the FX5200 doesn't do FP32 in hardware? (i was in vacation so surely i missed something)

For the 2nd part. I'm talking from Nv point of view not from board makers, i think it's clear from the beginning...

 

[Dave Baumann]

Anyway, where can i find the link where it shows that the FX5200 doesn't do FP32 in hardware? (i was in vacation so surely i missed something)

http://www.digit-life.com/articles2/ps-precision/index.html

For the 2nd part. I'm talking from Nv point of view not from board makers, i think it's clear from the beginning...

Yes, and your statement bears no relation to the figures being talked about here.

 

[digitalwanderer]

Question: What is this "waiver" y'all are talking about? Are you talking about M$ giving a free pass to some sub-WHQL drivers for nVidia?

If so, this is the first I've heard that M$ allows "waivers" to their quality certification program....I'd be VERY interested in hearing more.

 

[Tridam]

I've made many tests with a FX5600 and a FX5200. When using PS 2.0, I can't see any difference between these 2 chips. With 44.03, you can't use FP32 but with beta drivers or quadro drivers, this problem doesn't exist. FX5600 and 5200 hardware seem to do FP32 properly. The main difference I've seen between the two chips is that FX5600 can use its FX units to do PS1.4 but the FX5200 can't. FX5200 uses FP units for PS1.4.

I think the problem is not there. Does WHQL certification require DX9 compliance for drivers? I don't think it's the case.

Does Microsoft say "this chip is DX9 compliant" or "this one isn't"? I don't think it's the case.

If an IHV says that its drivers are WHQL, does that mean that they are DX9 compliant?

If an IHV says that its new chip is DX9, does Microsoft ensure for compliance? Isn't it a choice of the IHV to presents its products as a DX7, DX8 or DX9 parts with a single constraint: not shock press and people to much??? I've never seen a clear Microsoft's definition of a DXx product.

 

[indio]

All I know is that if you start googling the names of the 2 people that own Mercury Research , Dean Mccarron and Michael Feibus, you will find they are contributing editors of Microprocessor Report

http://www.instat.com/images7/analysts/peter_glas.gif
remember this guy when he gave the NV30 the 2002 processor of the year award?
Peter Glaskowsky


I think it's been well established that Microprocessor Report is a PAID analyst. I've also found that Mr Glaskowsky and Dean McCarron are often quoted together in the same "news" stories. I'm sure this is done for the purpose of making the story "appear" as if it queried multiple independant and diverse experts
Mercury Research is trash to me. In light of the corporate events of the last 18 monthes . All analysts are HIGHLY suspect

 

[digitalwanderer]

All I know is that if you start googling the names of the 2 people that own Mercury Research , Dean Mccarron and Michael Feibus, you will find they are contributing editors of Microprocessor Report

http://www.instat.com/images7/analysts/peter_glas.gif
remember this guy when he gave the NV30 the 2002 processor of the year award?
Peter Glaskowsky


I think it's been well established that Microprocessor Report is a PAID analyst. I've also found that Mr Glaskowsky and Dean McCarron are often quoted together in the same "news" stories. I'm sure this is done for the purpose of making the story "appear" as if it queried multiple independant and diverse experts
Mercury Research is trash to me. In light of the corporate events of the last 18 monthes . All analysts are HIGHLY suspect

I'm not disagreeing with you and don't mean it as fighting words, but what exactly does this have to do with the discussion? How does Mercury fit in on this one? (And does this mean I'll get to play with me friend Peter Glaskowhatshisface again?!?!? OH BOY! I think I almost made his head explode last time... )

 

[RussSchultz]

Ok, demalion. One final try.

I feel a qualified reviewer should know either that it does, or does not have whatever quality he is claiming it has (or has not). Probably shouldn't enter the question, in my opinion.

Even if the quality in question is "the ability to run dx9 games at playable framerates" (which, I'll have to disagree that the construct of the sentence necessarily means that, though yes its possible that it that is what he meant) he should have enough familiarity with the product after reviewing it to know with more certainty than "probably".

But, assuming we take properly to include "at high enough framerate to be acceptable":

Truly, it is a bit murky to know what a DX9 game will require in terms of performance, however In my opinion, any reviewer worth his salt wouldn't come to this conclusion especially after just having run the card through a shootout and demonstrating and concluding that the card does just fine on current games at a lower resolution. Assuming it has the same feature set as its larger brother (which it does), it should accomplish the identical thing on future games. Or, are we now to assume that properly means "at a given frame rate at a high enough resolution" so that once again I am "wrong"?

Beyond that, given that the bulk of DX9 cards sold are indeed these cards which would suggest, in my opinion, that they would have the effect of having DX9 games written in ways to make them playable so that there will be a large enough audience to sell to, again undermining the notion that they will be unable to "properly" play the games. And yes, in my opinion, a qualified reviewer would come to a similar conclusion, given the historical evidence that "no card gets left behind, ever" and the general idea that publishers want the largest audience possible.

All in all, the statement comes off, in my opinion, as either unfamiliarity with the product, unfamiliarity with the market, or predictive naysaying and dare I say, fanboyishness, particularly in context with the rest of the statement. (Essentially dismissing the 5200 as a DX9 card to invalidate the claim that NVIDIA has sold more DX9 cards). Which is breaks another one of my requirements of being a good reviewer--impartiality.

But, that's all my opinion.

 

[indio]

The number's quoted are derived from a Mercury Research quarterly report , although they do a good job not referencing it till the middle of the story. That could just be their writing style though.

edit : The numbers quoted by Xbit Labs that is.

 

[RussSchultz]

but what exactly does this have to do with the discussion? How does Mercury fit in on this one?

If you go read the article at xbit which is tangentially at the center of this conversation, you'll see they base their story on figures from Mercury Research. (Which, contrary to what indio says, is noted in the first sentence of the story)

 

[indio]

I'm sorry I was looking at this other story they did http://www.xbitlabs.com/news/video/display/20030731084600.html . My mistake :?