Nvidia/AMD/VIA leave Sysmark company Bapco

[Idontcare]

Good riddance. Tasks that computers run are increasingly diverse and rely on different components' respective abilities. I haven't even looked at SYSmark in years. I look at the benchmarks that more accurately reflect the user's specific needs. If they don't have specific needs (i.e. they are a general non-intensive internet/office user), I try to pitch an SSD and/or discrete GPU because the overall system usage experience is so much nicer than with a mechanical disk or integrated graphics.

CPUs just aren't as important as they used to be, at least not to the typical user.

I agree.

There's a rather smallish piece of the home consumer market segment which needs/uses workstation-caliber computing resources and then there's the other 95% of the market that would probably be fine with a Q6600/PhII class quad-core and an SSD.

This Sysmark bench, what percentage of office workers need anything remotely close to workstation-class desktop performance out of their CPU's?

Is power-point and email really that demanding? I don't think so, otherwise iPhone's would not be all the rage in businesses.

My wife has an iPhone through work, her work laptop HDD grinds and thrashes so badly when navigating the web that she's given up entirely on using the laptop to do anything internet related for her job. The pain threshold is just too high.

Benchmarks have never told the full story, that's why there's more than one of them in existence.

 

[Arkadrel]

Is Intel biasing Sysmark to favor Intel's products (voter fraud) or is Intel simply fielding the superior products and the other's are naturally going to lose out in the next benchmarking cycle?

1) Is Intel biasing Sysmark to favor Intel's products = likely yes
2) Is Intel fielding a superior product than the other's= likely yes

but case 1) has probably gotten so bad now that all the others decided to leave.

Also why do you guys assume its about the GPU portion? as others have said, it favors "score" wise certain elements more than others, which just happends to be the ones where Intel does good in (thats about the CPU).

Anyways... Intel owns the thing and is obviously useing the benchmark to make themselfs look good by "almost" cheating.

 

[Genx87]

Actually I thought I had made it obvious that this was the point of my post - that there are two sides, two possible reasons for this one outcome, and the problem is that we can't make any conclusions in terms of deciding the "why".

Is Intel biasing Sysmark to favor Intel's products (voter fraud) or is Intel simply fielding the superior products and the other's are naturally going to lose out in the next benchmarking cycle?

Didn't I address both of these possible progenitors in my post?

We can't take the "loser" at their word on this because they have a duty to themselves to always cloud the issue when the outcome is not to their favor, ergo we need (and so to does AMD/Via/NV) some kind of third-party auditing done to verify the claims.

I would have thought this much was just stating the obvious. Who knew that stating the obvious could generate any controversy whatsoever LOL

At any rate, hasn't AMD already been through this cycle once before, crying foul over optimized benchmarks? You'd think they'd learn already, get ahead of the 8-ball and formulate their own variant of Bapco sysmark.

Hopefully this time we'll see them take that step.

Yup, they tried to trash benchmarks that showed their inferior tesselator failing to get close to Nvidia. This is a pattern I expect over bulldozer. BD isnt going to stack up against SB much less IB. Now it is time to attack the benchmarks as flawed and biased.

 

[386DX]

Yup, they tried to trash benchmarks that showed their inferior tesselator failing to get close to Nvidia. This is a pattern I expect over bulldozer. BD isnt going to stack up against SB much less IB. Now it is time to attack the benchmarks as flawed and biased.

AMD's (and there fanbois) are hypocrites. On one hand (Llano/fusion) they are trying to tell you that CPU performance isn't important, even the slowest CPU is good enough to do everything. Then on the other hand (Bulldozer) they tell you that four cores isn't enough, you need 8 cores cause apparently everyone has Cinebench and 3dstudio on there computer and that represents Real World performance.

 

[busydude]

AMD's (and there fanbois) are hypocrites. On one hand (Llano/fusion) they are trying to tell you that CPU performance isn't important, even the slowest CPU is good enough to do everything. Then on the other hand (Bulldozer) they tell you that four cores isn't enough, you need 8 cores cause apparently everyone has Cinebench and 3dstudio on there computer and that represents Real World performance.

You are missing the context.

 

[GaiaHunter]

Yup, they tried to trash benchmarks that showed their inferior tesselator failing to get close to Nvidia. This is a pattern I expect over bulldozer. BD isnt going to stack up against SB much less IB. Now it is time to attack the benchmarks as flawed and biased.

And NVIDIA left just because they love AMD so much.

I loved how this turned into AMD/BD vs Intel/IB.

As a consumer I dislike synthetic benchmarks.

 

[Idontcare]

Yup, they tried to trash benchmarks that showed their inferior tesselator failing to get close to Nvidia. This is a pattern I expect over bulldozer. BD isnt going to stack up against SB much less IB. Now it is time to attack the benchmarks as flawed and biased.

Hypothetically speaking, if this were true I'd be more apt to give AMD/Via/NV props for finding their spines and going on the offensive in pro-active rather than reactive fashion :thumbsup:

I can respect predatory competitiveness, premeditated or not.

The persistent "whoa is me", forever the victim of circumstance and sinister ne'erdowells is what gets my goat.

So I say "good on you AMD, NV, and Via...now go do something about it!".

(and no, just as Larry Ellison of Oracle stated, lots and lots of blogging is not "doing something about it")

 

[Gigantopithecus]

AMD's (and there fanbois) are hypocrites. On one hand (Llano/fusion) they are trying to tell you that CPU performance isn't important, even the slowest CPU is good enough to do everything. Then on the other hand (Bulldozer) they tell you that four cores isn't enough, you need 8 cores cause apparently everyone has Cinebench and 3dstudio on there computer and that represents Real World performance.

You know, I hadn't thought about it like that. Your argument makes perfect sense because Llano and Bulldozer are aimed at the exact same market.

 

[jones377]

sysmark isn't a synthetic benchmark though. It seems 99% of people even around these parts don't know what a synthetic benchmark really is. sysmark is a (computer) system benchmark, hence the name.

Not taking a stand for or against sysmark. I never look much at those scores because while I may use some of the apps baked into it's score, I stopped being concerned about how they performed on my computer a few CPU generations ago. Even Acrobat seems to perform fine nowadays, which in itself is sort of a miracle. (not even sure if acrobat is included in sysmark nor do I care)

 

[Mopetar]

s
Not taking a stand for or against sysmark. I never look much at those scores because while I may use some of the apps baked into it's score, I stopped being concerned about how they performed on my computer a few CPU generations ago. Even Acrobat seems to perform fine nowadays, which in itself is sort of a miracle. (not even sure if acrobat is included in sysmark nor do I care)

The problem is that the scores do not present a meaningful representation of how well one CPU performs against another. If one CPU has a SYSmark score of 340 and another has a score of 170, it does not indicate that the first CPU is twice as fast as the second CPU. In and of itself, this makes it absolutely worthless for anything other than an absolute comparison, i.e. saying that one CPU is faster at completing some task than another CPU. You could interchange the scores with words like 'banana' and 'mongoose' and the result would only be slightly less useful.

I think Anand would be better off developing his own set of benchmarks for CPUs much like the benchmarks he created for testing SSDs. Such an approach might also produce interesting figures related to power use.

 

[Idontcare]

I went and dug up that old "benchmarks are optimized for Intel" article I was thinking of when I referenced it above:

If fact, the head of a competing benchmark firm bluntly labeled BAPCo as a "front" for Intel. Randall Kenney, the head of CSA Research, made this statement during his presentation at Platform Conference in early 2001.

http://www.vanshardware.com/reviews/2002/08/020822_AthlonXP2600/020822_AthlonXP2600.htm

It was actually Bapco, even then, all the way back in 2002.

Looks like AMD spent 10yrs trying to reform them of their pedigree (Intel) and finally just threw in the towel.

I look forward to whatever substitute they have planned. Hope this is a little more thought out than their curt dismissal of Dirk...otherwise we'll be still waiting for their proposed Sysmark2012 substitute in 2014...

 

[badb0y]

Yup, they tried to trash benchmarks that showed their inferior tesselator failing to get close to Nvidia. This is a pattern I expect over bulldozer. BD isnt going to stack up against SB much less IB. Now it is time to attack the benchmarks as flawed and biased.

LOL and what happened? No games have come out with so much tessellation that the HD 5000 series haven't been able to produce playable FPS.

 

[Phynaz]

And NVIDIA left just because they love AMD so much.

I loved how this turned into AMD/BD vs Intel/IB.

As a consumer I dislike synthetic benchmarks.

Except it isn't synthetic.

And Nvidia most likely left because they don't make x86 cpu's.

 

[Genx87]

LOL and what happened? No games have come out with so much tessellation that the HD 5000 series haven't been able to produce playable FPS.

Apparently AMD doesnt share your view of the situation. Otherwise they wouldnt had made such a big stink about it.

 

[Idontcare]

LOL and what happened? No games have come out with so much tessellation that the HD 5000 series haven't been able to produce playable FPS.

Kinda true of sysmark and office productivity metrics too...since when has powerpoint and email presented a challenge to modern desktops?

We all wait for the same thing when using office rigs - data being fetched from somewhere outside our computer. Be it local network access, or intranet, or internet.

KillerNic missed their market calling

 

[Phynaz]

The problem is that the scores do not present a meaningful representation of how well one CPU performs against another.

Why not?

 

[badb0y]

Apparently AMD doesnt share your view of the situation. Otherwise they wouldnt had made such a big stink about it.

It's called marketing. After being 6 months late to the competition and being only 10% faster than the HD 5870(I am referring to the GTX 480 launch) and consuming more power than a HD 5970(Again referring to the GTX 480), nVidia needed to market the video cards somehow to the masses.

 

[Vesku]

Why not?

Mainly because the scoring system inflates small differences in benchmark completion. This helps to sell the new stuff as WAY better than that "old crap" from a few years ago. However, the weighting of individual benchmarks appears to have been a contentious issue within BAPCo. As Idontcare referenced, the fight with Intel over Sysmark benchmarks and their relative weight for overall score has been going on for 10 years or so. Some companies finally decided it was no longer worth their time and money to be a part of it.

Similarly referencing Idontcare, they better have an alternative to promote. Otherwise they are just wasting everyone's time grumbling about how they got outmaneuvered in the Club President election and they're taking their ball and going home.

 

[Nemesis 1]

Makes sense that NVIDIA and AMD would leave if Sysmark is actually taking a step backwards and barely including GPU performance or excluding it entirely. May explain VIA too, they've been making some leaps in their IGP performance. But could also stem around compiler choice or perhaps nothing in Sysmark takes advantage of their on chip encryption acceleration?

Ya I can see AMD wanting to benchmark for fusion makes sense . But to want the benchmark befor apps using open CL has any real presense in the market . Seems like AMD doesn't want to be benched marked against IB . IB looks to be coming to market at the proper time. Maybe AMD can go to Steve Jobs and say . Hey look were ready with our fusion product with Open cl. Apple being the push behind open CL should jump all over that . BUT it just won't happen .

 

[StrangerGuy]

Well, it's not like anyone with a clue even cares about Sysmark.

 

[Genx87]

It's called marketing. After being 6 months late to the competition and being only 10% faster than the HD 5870(I am referring to the GTX 480 launch) and consuming more power than a HD 5970(Again referring to the GTX 480), nVidia needed to market the video cards somehow to the masses.

So AMD made a big stink about tesselation because Nvidia came to market 6 months late with a part being 10% faster?

Interesting theory.

 

[Nemesis 1]

You are missing the context.

No he isn't . The fact remains even tho Llanos igp is faster than SB IGP neither are good enough. You guys always dig yourelves a hole and later double talk . When IB beats llano in gpu benchmarks. It will be another tune sang by the green team as they pledge allegiance to trinity. We all know this to be a fact and true.

 

[badb0y]

So AMD made a big stink about tesselation because Nvidia came to market 6 months late with a part being 10% faster?

Interesting theory.

Do you read? Tessellation was nVidia's strong point so they pushed it as hard as they can in synthetic benchmarks such as Stone Giant and Heaven. AMD made a big stink because of nVidia's marketing trying to sell their cards based on synthetic benchmarks rather than real world performance.

 

[iCyborg]

No he isn't . The fact remains even tho Llanos igp is faster than SB IGP neither are good enough. You guys always dig yourelves a hole and later double talk . When IB beats llano in gpu benchmarks. It will be another tune sang by the green team as they pledge allegiance to trinity. We all know this to be a fact and true.

Good enough for what? Gaming at high-res? Sure, but the MMO crowd with 1366x768 notebooks will be served well.
It doesn't look like IB will beat Llano in GPU, and even worse is that given the rumoured March 2012 launch it will go against Trinity.

 

[Soleron]

IDC is arguing with himself now. This should be good.