Kyro II Beating Geforce Ultra Again

[dfloyd]

Planet Hardware



Just in case the one review that was posted below wasnt enough.


Oh and to all those who posted advice on my Kyro II vs Geforce Pro vs Geforce III etc, Thanks. I have made up my mind and right now it doesnt look like the Kyro II can be beat in a price/performance comparison. Not to mention I am getting 256MB more ram (Total 512) and a Hercules Game Theater Xp (thats the video card, sound card, and extra ram for less than just the G3 would of been).

Yeah the G3 is faster and does support more fancy features. But when I really look at the G3 I cant help but think of the Geforce I. Yeah it had T&L, but did it ever USE T&L? It is true that the G3 has all the features but I think there will be two more iterations out before we see any software that really takes advantage of it. So on that note I decided not to go with the G3. Not to mention the very high price.

Thanks again though to all who posted.

 

[Leon]

The guys at PlanetHardware obviously didn't know how to configure their systems right....

PlanetHardware MDK2 scores

Anandtech MDK2 scores

I can't help but laugh.

Leon

 

[ThExorcist]

Glad the advice could help...

thanx for the review link

 

[richleader]

Yeah, but microsoft wasn't building a game rig based on the Geforce 1...

 

[dfloyd]

You are correct and the problem with that train of thought is I absolutely hate the X-Box. It can do nothing but prevent more games from comeing to the PC. Yes they should be easy to PORT (God Forbid). But we all know how much time companies spend on Consol to PC Ports.

The PC is my system of choice for gaming. Even if it did not have the best sound, best graphics, best whatever it would still be so.

The reason: Nothing offers as much expandability as the PC.

Look at it from the Mod perspective. Without the PC there would never have been a Team Fortress, Counterstrike, etc. Also consider the extra maps, levels, and other stuff you can get because of your pcs expandability. Sorry but the Xbox hurts the PC like no device before because it comes closer to being a pc but still lacks the pcs expandability. So SHORT LIVE THE XBOX! (I know it will be a hit but I can always hope)

Sorry for the rant and yes I see your point but it still does not justify that much for the video card. Yes the cards features will be used in the Xbox software, how long before we see any ports of the xbox software to pc, and how long before there are good ones done (see above). So your still stuck in two or three iterations away before (At least the second iteration of the G3 as by September when the Xbox is released I am sure the G3 Ultra will be rounding the corner) any games really take advantage of it.

 

[richleader]

And I'm just saying is that there's a million contradictory benchmarks and that the average person is better off spending $130 for a geforce 2 pro. If Kryo 2 floats you boat, great, people like you who fall for alternatives make the real hardware cheaper for the rest of us. Just messing with ya, but in all honesty, I don't think the installed base of Kryo 2 users is large enough to be positive about anything concerning the product--especially as it's most ardent defenders here, are beyond rabid fanatics. I wouldn't buy a Geforce 3 just yet, either, but if you have an MX/radeon-le card now, the Kryo II isn't a smart upgrade.

 

[Taz4158]

<< And I'm just saying is that there's a million contradictory benchmarks and that the average person is better off spending $130 for a geforce 2 pro. If Kryo 2 floats you boat, great, people like you who fall for alternatives make the real hardware cheaper for the rest of us. Just messing with ya, but in all honesty, I don't think the installed base of Kryo 2 users is large enough to be positive about anything concerning the product--especially as it's most ardent defenders here, are beyond rabid fanatics. I wouldn't buy a Geforce 3 just yet, either, but if you have an MX/radeon-le card now, the Kryo II isn't a smart upgrade. >>


Hehe. The Kyro people are &quot;rather zealous&quot; aren't they. I don't recall us early adopters of the Radeon being quite so frenetic or hyper.

 

[PotNoodle]

? but in all honesty, I don't think the installed base of Kryo 2 users is large enough to be positive about anything concerning the product?

What on earth is that supposed to mean?

However, considering this is a completely new product I think you?ll find the installed user base is growing pretty damned rapidly; ask Hercules, I?m sure they?ll concur.

?especially as it's most ardent defenders here, are beyond rabid fanatics?

Gee, thanks ? point out something that I;ve said that isn?t correct.

Of course that accusation could just as easily be leveled at many other people here as well.

?either, but if you have an MX/radeon-le card now, the Kryo II isn't a smart upgrade.?

Now that I do not understand - its clear to see that KYROII is a good upgrade from an MX under most circumstances ? had you said GTS / Radeon or possibly a V5 then I would concur; MX? Its an upgrade.

 

[richleader]

It's an upgrade. But not large enough that anyone in their right mind should shell out cash for it--that money would be better utilized on a full directx 8 part in the future--but if someone has money to burn, who am I to judge.

Benchmarks for the Kryo 2 have yet to stabilize to the degree that Nvidia and ATI cards have, as is evident between the comical difference between the two marks given by Leon here. I suspect that Planethardware disabled the use of T&amp;L (as many reviewers did/didn't do to compare the geforce 2 to the Voodoo 5), but that's only a guess.

 

[Taz4158]

<< It's an upgrade. But not large enough that anyone in their right mind should shell out cash for it--that money would be better utilized on a full directx 8 part in the future--but if someone has money to burn, who am I to judge.

Benchmarks for the Kryo 2 have yet to stabilize to the degree that Nvidia and ATI cards have, as is evident between the comical difference between the two marks given by Leon here. I suspect that Planethardware disabled the use of T&amp;L (as many reviewers did/didn't do to compare the geforce 2 to the Voodoo 5), but that's only a guess. >>


To quote a popular TV show....the Kyro isn't quite ready for primetime. I grabbed one on the weekend to &quot;test&quot; out and dutifully formatted and installed it and found it's little more than a toy. Can't compare at ALL to a 3 or even a Pro. Frankly I don't play MBTR and don't know anyone who does.

 

[PotNoodle]

richleader,

?It's an upgrade. But not large enough that anyone in their right mind should shell out cash for it--that money would be better utilized on a full directx 8 part in the future--but if someone has money to burn, who am I to judge.?

So, you?re telling us that an MX to a GTS (Plus in many cases) isn?t a worthy upgrade then? Its pretty easy to see that in most cases that with KYROII we are looking at GTS level performance, and in some cases (as these reviews are showing), beyond. If MX to GTS level performance isn?t an upgrade I don?t know what is.

? Benchmarks for the Kryo 2 have yet to stabilize to the degree that Nvidia and ATI cards have, as is evident between the comical difference between the two marks given by Leon here.?

Errrrm that statement has crossed purposes ? Leon comment was to point out differences in the NVIDIA scores, so in this case it?s the NVIDIA scores that haven?t stabilized based on the evidence you are providing!!!

However, in general terms, no of course KYRO hasn?t stabilized yet; it?s a new chipset and the work on the drivers has only had a fifth of the time to show the maturity NVIDIA?s detonators have. We?ve already seen numerous performance gains from KYRO?s drivers (the large leap in NFSU score from the last drivers being an example) and I fully expect to see even more.

? I suspect that Planethardware disabled the use of T&amp;L (as many reviewers did/didn't do to compare the geforce 2 to the Voodoo 5), but that's only a guess.

My PIII 733 stabilizes at around 35FPS with Hardware T&amp;L enabled on a non T&amp;L board (V5 and KYROII); the first link Leon Points to shows the GF3 attaining 80FPS on a 1GHz Athlon and I doubt that would have been achieved without T&amp;L on that CPU. Why the Ultra is just over half the speed of the GF3, though, I can?t imagine.


Taz4158,

? I grabbed one on the weekend to &quot;test&quot; out and dutifully formatted and installed it and found it's little more than a toy?

I was under the impression that?s all any 3D accelerator is.

?Can't compare at ALL to a 3 or even a Pro.?

But It can be compared to a GTS.

Personally I don?t compare KYRO to a Pro, and its not even supposed to touch a GF3 (being a third of the price!); where KYRO is achieving this level of performance is the exception rather then the rule, and it only does it in a few circumstances. However, its nice when you purchase a card to find that it occasionally does have performance as good as cards far beyond its price.

 

[Taz4158]

<< richleader,


?Can't compare at ALL to a 3 or even a Pro.?

But It can be compared to a GTS.

Personally I don?t compare KYRO to a Pro, and its not even supposed to touch a GF3 (being a third of the price!); where KYRO is achieving this level of performance is the exception rather then the rule, and it only does it in a few circumstances. However, its nice when you purchase a card to find that it occasionally does have performance as good as cards far beyond its price. >>



Keyword occasionally. It does well in some specialized benchmarks that would be of no use to me nor I suspect to the vast majority of people here. It of course does extremely well in Villagemark for what little use that is. Glad you're happy with the card but there are too many hyperbolous posts expounding the Kyro and really not taking into account the overall performance which is good for a BUDGET card and nothing more.

 

[nam ng]

1600x1200x32

Anantech old driver
slowpoked GTS 64MB ---> 50.4FPS
Lightning KyroII -----------> 41.4FPS

Planethardware improved driver
Hella fast GF2_ULTRA -> 40.8FPS
Lightning KyroII -----------> 44.7FPS

 

[jbirney]

Actually Leon,

it looks like Anandtech dose not know how to configure.... Check this from an Nvidia fan site:

RivaStation

Agreed that they do not have an Ultra. But notice how for a simular CPU the K2 and the MX scores are right on the money?

I thought that planethardware said that they had issue with the Ultra. So I don't really take the Ultra numbers to count much for the most part. The ultra numbers seem to be right on in Q3. However do yo see how the K2 is doing over all? Not bad for a 64mb card going for $98.00 eh?

 

[dfloyd]

Ok only under certain situiations? Which situiations? From the benchmarks I have seen the Kryo II performs well under most all tests and does 1024 x 768 x 32 (at high quality settings) with ease.

Also if you look at the benchmark it is using Nividas latest drivers. How much tweaking has to be done before Nivida gets better performance? Also if you notice on many of those benchmarks the G3 is just beating the G2Ultra (As it should be in most non dx8 games). So in that instance either he configured both cards wrong or the numbers are correct.

One thing I did notice is that he is using Win2k. Its possible Kyro does better under Win2k than Nivida cards. I dont know myself but just a possibility. I notice Anand used Win98se. If this is the case it further justifys my decision as WinXp should perform more like 2k than like 98. Just theorizing here but that seems to be the case.

 

[dfloyd]

Oh and I am not looking at the card as an upgrade from another. Rather a new card in a new system.

And I have to disagree with the person above who said you cant compare Geforce III to the Kyro II. As I did compare them.

And the Kyro II won. Not because it was fastest or had the most features but because it provided the most bang for my buck.

 

[loosbrew]

ivge had my Kyro II for a few weeks now and i cant be happier. it performs wonderfully in seriou sam with 110 FPS without FSAA on, and i hit usually ~70 with 2x FSAA on. image quality has been more than great and i havent run into any driver issues or incompatibilties at all yet.

loosbrew

 

[PointlesS]

for the people who are wondering why Kyro II is getting so much attention....I think it's because it's new technology (well...not really because it was used in the old powervr chipsets....but it's brought back to life) the gf2 has been around for longer than a year now and everyone knows what it can and can't do and it's nothing new...Kyro II has lots of features for a sub $150 card and I guess this is what sparks an interest in many people (including me...) I wonder how it will perform in future tests when direct x 8.1 is out...

 

[dfloyd]

Thats the beauty of it. At $100 or so price tag (OEM from Pricewatch) you can afford to upgrade in six months. Which is what used to be expected to play new games. So when all the new Dx8 titles are out, so too will be the new Radeon, hopefully the new Kryo and probably the new G3 Ultra (Or pro or whatever it will be called).

 

[Archknight]

I think most of the responses from this forum are positives so far.
Anandtech review was specified that T&amp;L is enabled and we all know about T&amp;L in MDK2. I would assume that planethardware had the T&amp;L disabled. But then again would disabling T&amp;L give the Kyro2 higher fps since it is not that CPU dependent.

One thing for sure, Kyro2's driver is getting better and better.

Maybe the older version driver used by Anandtech had improved over the time period and the result is planethardware results. Maybe.

 

[merlocka]

From the Planethardware review

First
Our GeForce2 Ultra turned in some lower-than-expected scores in this benchmark, and something seems fishy at the lower resolutions

Then
The Kyro II is definitely flexing some muscle at the higher resolutions, besting the GeForce2 Ultra handily.

Their review is bogus, I think they have some configuration problems.

For example, not only does the Kyro spank the Ultra, but so does an Geforce2 MX in the Expendable testing 1600x1200x32. Hmmm.

Expendable 10x7x32 P4 1.7GHz
Ultra ------ 76.32fps
MX400 --- 86.57fps

Also, the MX performance at 1024x768x32 in Q3 decreases 20% buy putting it in their AMDS setup.

Quake3 A 10x7x32 Geforce MX400
Athlon 1GHz ---------- 41.9fps
Pentium4 1.7GHz --- 52.4fps

The kyro2 is a great card, good technology, etc. But don't buy one cause of this &quot;glowing&quot; review from PH. I think they have a bad setup.

 

[dfloyd]

They may have a bad setup. But I do not think so. Compare the Geforce III benchmarks. Almost the exact same as on Ace's Hardware. Compare the other benchmarks to many other sites and you will see that many are right on par.

And is it impossible to drop 20% when going from a 1.7GHZ processor down to a 1GHZ?

And under just about every test I have seen the Geforce cards do perform slower under Win2k than under 95/98. Not to mention they seem to perform better with Intel processors.

 

[merlocka]

I never said the Geforce 3 numbers were unexpected.

The Geforce2 Ultra numbers are unexpected and don't jive with scores on any site I've read. They saw the same discreptancy, but printed the article and said the Kyro beats the Ultra handily. That's not good journalism.

They should have considered that their audience would have read most of the other articles on these cards and exlpained why their data was different. Instead they call their own measured results &quot;fishy&quot; and go on to make a conclusion from that data.

>And is it impossible to drop 20% when going from a 1.7GHZ processor down to a 1GHZ?
In ths case, yes. The MX is fillrate limited at about 50fps in this specific test.

I'll ask you a question. Why cant a 1GHz AMD Athlon push the card to it's fillrate? Perhaps because their AMD test bed or Detonator drivers were not funtioning properly for the tests.

...and as Leon has pointed out, the Anandtech data shows the Ultra at 50% greater performance that the PH data.

Again, I'm not saying anything about the Kyro. I'm just questioning the validity of the data in the PH review - specifically, the data on the Geforce2 Ultra.

 

[Remedy]

Why do you guys read PH in the first place? They're just as bad as Sharkey and FiringSquard(god only know who visits FS). Those sites have poor management and editors to go along with it. Do you think it was an accident that the conclusion came out to be what it is?

 

[dfloyd]

No but I honestly think they reported the numbers they got. If you think this review was biased you have obvisouly not read much from Tom.

As far as not going to these sources. I totally disagree. Get all the info you can from anywhere and then compare it all. I compared many of the benchmarks to other sites and they are right on. Guess they screwed up by getting the same fps that 20 other sites did. Yes some numbers are low but I honestly believe thats what they got. And if someone cant install a video card and the newest drivers correctly I am sure they would not be doing reviews for a big site like Planet Hardware.