Slashdot Mirror
Trident XP4 Reviewed
ceebABC writes "In a new review, the Trident XP4 got a nasty reception. Based on the tests, it sounds like Trident has got some work to do on the thing. Looks like this GPU is dead on arrival." Our last story on Trident mentioned them coming back from the dead. Maybe not.
...And this is why you shouldn't believe the prerelease specs. Nothing ever performs to spec; trust the benchmarks.
It's pretty common, really. But I haven't usually seen it to this degree.
Warning: Poster of this comment is a nerd. Just like everybody else here.
That review
has really
great formating.
I just love
to read in
one thin
column. Or
maybe they just
have funky
formating for
IE?
The surprise isn't how often we make bad choices; the surprise is how seldom they defeat us.
than my lowly i810. :)
P.S. Am I the only one getting connection issues on Slashdot? Has Slashdot been Slashdotted?
"Times have not become more violent. They have just become more televised."
-Marilyn Manson
This card rivals the ATI Raedon for the 'worst drivers in the industry' award?
It makes me wonder why an AnandTech article gave such a different opinion. Which one is right?
From page two...
They only test one resolution, 1600*1200! Maybe it's just me, but I don't see a lot of laptops with 1600* resolution. The whole review is only meant to make the card look bad, it doesn't take into consideration price, power/heat consumption, or other important factors. It is biased, shallow and not worthy of a /.ing!
Moderation: Put your hand inside the puppet head!
It would be interesting to see a review of the card at normal resolution (the target market for the Trident probably can't even do 1600x1200 on their monitor, 1024x768 is a more reasonable resolution), and comparing it to a typical two year old card.
If it does hardware T&L and doesn't cost much, it would be a nice replacement for the ATI Rage 128 Pro that I have.
Hardly. The product reviewed was far from the polished final version we'll see in stores, and the drivers were beta and buggy. I'm not saying it will live up the "80% of a Ti4600" claim, but the price point will put it in competition with the vastly inferior MX series.
Regardless, Trident's biggest customer has always been OEM's, so if they can deliver a cheap, decent card, they'll easily hit their target market.
The future isn't what it used to be.
The review starts off saying this is a GPU for $100 cards and then compares it to GF4-4200 and ATI9500 Pro. Then proceeds to laugh at it for poor comparisons. Methinks Trident is going to laugh all the way to the bank when they clean up the cheap prebuilt box with embedded video market.
back when I was poor(college) I bought my first PCI card for $45. It was a trident 1Meg upgradeable to 2. It worked good for Doom. Now most of the new kit hitting the market is pure, unadulterated junk and it costs more. I fear this trend is related to the overall decline in the tech economy. I think I will hold out on my purchases until these companies find a way to put some cash back into thier R&D bugets and increase the quality of thier products. Maybe cut CEO salaries?
Karma: Censored (mostly affected by decency laws)
(first, a note to story submitters - when possible, link to the printable link version of the story. It is SO much nicer to read.)
Ouchie! Man, I wonder if Trident will EVER let these guys review ANYTHING again!
I hope, for the sake of the engineers at Trident, that there was some major D'OH! in the code, and that this isn't where their product really falls.
www.eFax.com are spammers
... from the lack of posts. Don't you know we're oly interested in Microsoft's failed projects?
P.S.: Anyone else experiencing extreme sluggishness about /. today? Earlier, I've had articles loading in background tabs for more than a minute. o_o
P.S. Update (10 mins later)!: Ouch. Maybe it wasn't such a good idea, noting the above comment, to intoduce another step and try previewing before submitting. -_-
That was a pretty bad review.. Matrox tried to "clean up the floor with nVidia and ATI" too and failed. At least they made a better attempt than Trident.
I have a spare 2mb Trident in my P-75 that may compete with the XP4...
I admit that I don't know much about graphics cards or GPUs, and it's obvious that the Trident got smoked. But was it a fair comparison? According to the review, the Trident goes for under $100, yet they benchmark it against what appears to be higher end (more expensive) graphics cards.
Wouldn't it be a more fair comparison if they benchmarked against cards of the same price range? If you were shopping for a cheap card, you obviously wouldn't expect it to perform as well as a more expensive card anyway, would you? What do others think?
---
Open Source Shirts
Historically, for the last 6-7 years, Trident has always focused on the mobile graphics market, and in that space, they are much more dominant. The XP4 is basically an evolution of Trident's mobile GPUs, and is really intended for use in mobile systems, hence the considerations such as reduced transistor count, etc. There's little difference between the mobile XP4 and the desktop XP4, and yet Trident is marketing it as a desktop one.
For a laptop, the 3D benchmark scores are actually quite decent.
But for them to call it a desktop GPU is just asking for trouble, as the article clearly describes.
There's 10 types of people in this world, those who understand binary and those who don't.
Then he proceeds to run this card at 1600x1200 with beta drivers against cards with excellent debugged drivers. Any sort of numerical or empirical evidence he could get at this point is about .05% useful to me.
Sure the thing may only get 4.9 FPS on a new demanding game at 1600x1200 with beta drivers. I bet you that same card will belt out over 60fps at 1024x768 when the real drivers are released.
People seem to forget that a video card driver's quality can be the difference between horrible performance and class leading performance. If the driver is not debugged and performance optimzed, there is nothing a hardware designer could do to make that card perform well.
I say that this is an excellent card that will allow users who do not want to spend $500 on a video card to play the latest and greatest games on the market. It is a Dx9 card, with full support. To me, this is an excellent card.
I bet they sell a whole boatload of these things to OEM manufacturers and those who do not really want to spend an entire car payment on moving some pixels around. -TinyManCan
PR: It has come to our attention that many of our customers and critics are not satisfied with the review of our product as we previously shipped to them and received with deafed ears. The staff of www.extremetech.com have mis-interpreted Trident's XP4 product and have mis-applied our technology. In our initial PR Announcment of the XP4, we were received by listeners that our product will whipe-out the competition. Despite our best efforts to contact the staff of www.extremetech.com before they released the results, we have received much criticism and have now been given opportunity to make clear our statements. Our initial PR statment confirms that our product was not intended to whipe-out the competition; we meant that the XP4 will whipe the ass of our competitors. We understand the definite language barrier of our PR staff and the general international public. Over the past 6 months, Trident has become one of the greatest suppliers of industrial sand paper and the most abrasive toilette paper in the history of indoor plumbing of developed nations. The Trident XP4 is intended to provide the most dis-comfort in our competitors as its only use is to whipe their ass in the most abrasive fassion possible. We thankyou for your concern and please feel free to purchase more of Trident's innovative products.
I thought this information should be re-layed to the slashdot community as it clears-up much of the incorrectly perceived statements. You know what happens with the SNAFU theorom these days...
Sincerily,
Bob Grover
I failed to see any testing of its performance doing DX9 specific tasks. It obviously isn't going to smoke a GeforceFX card, but will it be better than a Geforce3 or Geforce4 at running DX9 and OpenGL 2.0 shaders?
And I would have really liked to have seen them run the tests at 1024x768 anyway despite the lack of AA in the drivers.
I'm a loser baby, so why don't you kill me.
Ok one thing I didn't see (and IRTFA), was the overall "playability" of the games or applications.
Dumb down the tests and give it to joe-six pack (you know, the ones who WON'T spend the extra 300 bucks for 30 trillion pixal shading?) and see what they think.
Does it run the app fine? Does the game run smooth in a comfortable screensize?
Being broke lately, I've come to appreciate that UT2003, or Dungeon Seige runs just fine on my celeron 533 with 512 meg ram, and while a more powerful graphics card would make it run even better, my 2 year old Gforce2 works just fine.
Just fine for the Cheapo price I would pay for the same card nowadays.
Extremetech turned me off of readership in the past by their lack of credible articles, and this just reinforces why I stopped reading it.
Personal opinion should be available at the END of an article, not the beginning opening bias.
Well my Word document just decided to unfreeze and let me save, so I will end this rant.
Yo Grark
Canadian Bred with American Buttering
Canadian Bred with American Buttering
They show the results of very different tests. 800x600 vs 1600x1200 resolution. That's a difference of 4x in required fill rate. Since neither of these tests were run at a relevant resolution (most laptops run around 1024x768), neither can be called conclusive. My guess it that Anand is waiting for more stable drivers to test "real" resolutions.
The extremetech.com review is pretty unfair, it's like testing a new Ferrari by seeing how much cargo it can carry and then declaring it a bad car because it doesn't haul as much as a Ford Explorer. This card is aimed at lower resolution (lower fill rate) applications that require low power and cost. Having a DX9 entrant into this arena to me is welcome.
We'll just have to wait for a real review to see if this card is any good.
-Ryan C.
-Ryan C.
It seems that Trident want's back into the GPU mainstream. They developed this card, the XP4, and are releasing it to benchmarkers early for reviewing purposes as many card makers do.
In one review, we have extremetech maxing up the resolutions and detail levels of some heavy hitting games, in addition to a 3dmark benchmark, against two of the biggest cards out there. These cards are at least twice the MSRP of this card. Extremetech then complains that the inexpensive card with beta drivers doesn't tread water against the established champs.
In a different review, anandtech set the resolution to something normal (how many gamers out there actually run the game at 1600x1200?) and they show the card as giving fluid performance, even beating the Radeon 9000 in one map. Albeit still behind the other two cards reviewed on some tests, they do mention that the drivers are beta and that finalized they will probably make the card perform much better.
I've been noticing that extremetech's reviews seem really, well, extreme. At least from my perception they will give good reviews to what can keep pace with the top cards or exceed the top card - and at times seems to focus on the war between NVidia and ATI for the title of Supreme cardmaker.
But how long ago was it that both of these companies were in Trident's situation? How long ago was it that these companies were struggling against 3dfx?
Like many before me ahve said, wait and see. This card could turn out to be the best card price for performance wise. It could come out and have the mobile version do everything else in. It could come out and be complete crap against whatever new cards the twin titans come out with.
----- I want my LART.
Doesn't anyone else notice that the nvidia geforce chipset 3D demos are a bit misleading. Sure, those things look awesome, and even better, they're in real time!
The problem is, you won't even be able to see anything like that in a game anyways because there are more objects showing on the screen in a game. Heck, I bet a card that's 2 year older can pull something off like those demos with good graphics coding.
I just wish they would show something more practical.
Remember, this part is probably OEM targeted, not enthusiast marketed. Most users will say, "Gee, that thingamabob's got 2.4 gigahertz of RAM, wow it's fast!" and buy it, not realizing they got shafted on the video.
Carry out this philosophy across the machine, and you can shave $100-200 off the price of the machine, at least.
InThane
Well, considering how Trident was talking up how these cards would get blown away by their new graphics card, I think the comparison is wholy fair.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Judging from the benchmarks, including the fill rate, I think there's probably something terminally wrong with the current silicon or drivers. Hopefully this will be worked out before the product ships.
I say this because it seems odd that a card running at a reasonable clock speed with reasonably fast ram should run so slowly on fill-rate tests.
It's been a long time.
I dunno, 4 out of 5?
For all the issues with XP4 driver performance today, I have no doubt the Trident folks will improve it just like all other competitors in the past ....
Although the XP4 T3 with 128MBytes is listed at $99 MSRP, I would venture to guess that you will be able to buy it next month for less than $85, and next May 2003 for less than $60 (still with 128MBytes of course).
Trident's main value to MOST of us (working consumers with LIMITED income) is their RELENTLESS competitive drive to bring down the cost of high-quality graphics (which so far only affordable to the elite few) starting with DX8.1 today and DX9.0 next summer.
So here is my cautious forecast ....
Next August 2003, Trident offering will include a full DX9.0 card with 256MBytes XP8 for $99 and a full DX8.1 card with 128MBytes XP4 for $59 - all will support the latest AGP-8X of course.
This is why (I think) the OEMs, system integrators, retailers and distributors are so excited about Trident return to the desktop market.
As a final note, in case you did not know ....
It was a mere 45 days ago that the ATI 9000 Pro was selling for $149 with 128MBytes and everyone thought it was a GREAT price ! With Trident XP4 entering the 128Mbytes market with an incredible $99 MSRP, the 9000 Pro price has quickly dropped 33% or more to $99 and below !
I firmly believe that for us consumers, the surest way to get "the best deal in town" is by encouring more intense competition and not by killing it ...
Despite what their marketing might say, Trident is obviously not REALLY trying to compete with the latest and greatest from nVidia and ATI. This is obviously piece of budget hardware for people who don't want to pay a whole lot to get DX 9 compatible card.
Sure, if I build a gaming rig, this isn't the card I'm going to use. I'm going to spend the cash for a high end card, and probably brag about my insane frame rates the next time I take it to a LAN party.
On the other hand, if I built a PC for someone who isn't planning on playing Doom 3 extensively, I might actually consider a card of this calibur. It is a DX 9 card for under $100. This is probably a decent choice for a bargain PC.
"You spoony bard!" -Tellah
There was no point comparing a card targetted at the sub-100 market against boards in the $400-500 market.
Budget cards sell to budget markets, which means a 17" monitor that will do 1280x1024x75Hz with some degree of acceptability. Testing performance at 1600x1200 was pointless for this market.
This chipset is designed for a market where the whole system (less monitor) is selling for prices comparable to top of the line NVidia and ATI cards. It's not intended to compete with those cards, but to provide a tolerable experience on a cheap system.
I do not fail; I succeed at finding out what does not work.
why are they benching it against Desktop GPU's, obviously no matter what the laptop GPU the equivilent class GPU would always be better. Why don't you atleast bench it against mobile GPUs.
The boards being compared are in the roughly $100 market. Unless the Trident chipset hits the sub $50 integrator market, they're DOA.
I do not fail; I succeed at finding out what does not work.
While most of the /.ers have noted that the review of the XP4 on extremetech was unfair, they may not realize just how unreasonable extremetech was in putting this sub-$100 card against the best cards on the market.
Because the XP4 deviates from the long-established, direct-mode rendering (which is a brute force method) for tile-based rendering, they are going to need a lot of time to get their drivers in order before they will be able to compete properly with the familiar video cards. The only other card mainstream card that attempted this rendering approach was the Kyro series, which demonstrated that tile-based rendering does have huge potential and that drivers will make or break the card's performance.
Interesting enough, because video cards using the tile-based rendering method are more efficient by 200-300%* when compared to cards using the traditional method, they should see a much lower performance decrease as the screen resolution is increased when compared against direct-mode renderers (e.g. NVidia NV9 cards and ATI Radeon 9500s). While it's true that fill rates do increase substantially with increased resolution, direct-mode renderers simply will experience that much more overfill.
*Direct-mode renderers have an overfill rate of about 2 or 3; this means that for every pixel visible two or three more have been rendered and then disgarded. Tile-based renderers, on the other hand, disgard everything that won't be visible first and only render what's left, giving them an overfill of 0. Figuring out what to cull first before rendering has begun is more complicated than culling excess pixels after they are rendered; this complexity is what makes writing the drivers for a tile-based renderer such a difficult task.
Trident has set for themselves an incredibly difficult challenge: 1) Make a card that uses a tile-based rendering method, which means throwing out nearly everything the graphics card industry has learned the past couple decades. 2) In addition to the first task, they have added the complexity of sharing graphical resources, thus adding all the timing problems associated with such a configuration. If they achieve only 50% of the performance of Nvidia's Geforce4 TI4600, that alone would be a considerable achievement. If Trident meets the 80% performance target they set for themselves, it will be all the more impressive.
What really matters is what chipset would the OEM have used if this cheapo chipset wasn't arround. If the OEM are the really cheap ones, they will probably use this to replace the old savage chipsets and the likes. If it's a "second-cheapest" OEM they will probably be using ATI bottom line chipsets.
So, people buying the cheapest motherboards on earth are benefiting. And that includes many of my friends or people I know.
Yes, they can try to buy a separate ATI card, or go for a better motherboard with more decent card, but at least you have a basic right there in your cheapo initial buy. If you never make it to buy the radeon 9500, then at least you have something (that is in fact faster than ANY card of two years ago).
I mean, people that buy latest-greatest are paying $300 for the privilege of running the cards the rest of the world have two years ahead (Not saying it's not worth, but time is the key here, not crap/cool). At some point I payed extra to have it BEFORE (couldn't wait, Voodoo could really do things no other could do) but I do not feel so pressed now. So welcome back Trident!
unfinished: (adj.)
Since when does the mainstream computer user play games in 1600x1200? Even mid-range 17" monitors are going to get fuzzy in 1600x1200, 1280x1024 or 1024x768 is much more likely.
Firstly, the Trident people were idiots to give their cpu to a site like "ExtremeTech" whose focus is on x-treme desktop gamers. They should have stuck to "cheap laptop enthusiast" sites ;)
None the less:
You're comparing a MSRP of below $100 vs a "super special online lowest price ever street deal" of $100 for a $150 desktop GPU.
I can't find a *Retail* (vs OEM) price for the 9000 128 MB part below $150 USD here in Toronto.
AFAIAK the reviewer clearly pulled a dumbass move by running a grossly unfair comparison (the worst possible - 1600x1200x32bit-fsaa against a 9500 and a 4200 - how stupid is he), and of course is now being highly defensive.
Isn't the XP4 a *mobile* cpu? Does it use active cooling? Or passive? What about the Raedon 9000? So does that make them a good pair to compare?
I'd like to see what the retail storefront price and the bulk-OEM prices are of a MOBILE version of the 9000 vs the XP4 and the performance comparison at 1024x768x16bit once they've released their FCS tile drivers for the XP4.