Matrox on the verge of collapse?

No because they are a private company so no one knows. However, if they're laying off people it's obvious they aren't very profitable. Best case scenario they fired the people so they'd break even. If you're rolling in the dough, you don't need to fire people to cut costs...

 

Matrox simple "rested on their laurels". Their 2D products were top-notch for their time, and probably still is ATM, but the debut of all-in-one chips (with great 2D+3D) from NV and ATI simply meant one thing - who's gonna buy essentially-2D-only products anymore? I'm not even sure if their DV-editing products sold well.

Sad really... I went from a Millenium2+Voodoo1 to a Rendition Verite1000 and then back to the M2+V1... kept the M2 when I had dual-V2s and eventually (and reluctantly) pulled the M2 out when I had the V3 (reluctantly because I felt the M2 still beat the V3 in 2D).

 

I disagree. There are many companies laying off people although they're profitable. Banks, Microsoft, Intel, ...
If you lay off people when you already lose money, you're late. A company has to do profits, so laying off support staff is a good possiblity to maintain profitability without causing any damage in the long term, as support staff is easy to recruit again if the market conditions start getting better.

 

I disagree. There are some german magazines (c't, tec channel) which did signal quality tests. Parhelia's 2D quality was good but it wasn't any better than the one of an ATI or a quality NVIDIA board.

Further, they did not "rest on their laurels". G200 was a good 3D performer in comparison to the Rage3D, the Riva128 and the other pseudo 3d accelerators at that time. Further, the G400 GPU was the best thing available for several months.

 

It's kind of sad to even see a rumor of a company like Matrox go. Matrox was the original videocard manufacturer with over 26 years of making videocards .

Parhelia is a nice effort, and even might have been the top speed performer , had Matrox released it last year along with the Ti500 and Radeon 8500 like planned (some of you might remember my E3 report last year where I said I was really excited by the direction Matrox was heading and the Computer Games Magazine July 2001 issue , which stated that Matrox would release a "benchmark winner" before the end of last year. )

Right now, I don't know what to think . Matrox PR refuses to comment on opeds so, assuming the facts are true, it's just sad.

 

Nappe1: PNY handles all of the Quadro4 series cards.

Ingram Micro is the only place I've seen a Quadro4 NVS card, but they tend to sell to only people with business reseller licenses.

 

1. Link,pls. AFAIR Parhelia was better than any NV-product.

2. They did. Pseudo accelerators? What? G200 came out really buggy D3D-driver AND WITH NO WORKING OGL-driver! Matrox spent very long time to release the first one!

3. Further that G400's 'several months' are rather 'few months'...
On the other hand at the time the speed of G400/Max was on pair w/ TNT2 or slower...

 

Indeed... The visual differences (both scoped and display) at DarkCrow were quite visible (DVI output withstanding). Of course it also does have decent support TMDS output above 1600x1200 (for those of use with flat-panel displays with higher resolutions than that)...

 

Just a FYI...

There appears to be "some" Matrox info that's forthcoming. Stay tuned

 

Hey Typedef, would that be regarding their products or the state of the company?......

 

Found the link I mentioned:

http://www.tecchannel.de/tecdaten/s...ageid=989&sort=wertung&sortorder=desc

G550 is first, Hercules Radeon 8500 are 2nd and 3rd, Parhelia is 8th.

SiS boards have the worst 2D quality, followed by V5 5500 and various NVIDIA boards.

 

I couldn't read the article, but I looked at the graphs and Parhelia looked as good or better than anyone. What were their rating criteria?

Edit: typos.

 

For fairness sake the genuine SiS Xabre got a score of 6.5 (same as Matrox G450), with an Elite Xabre being the worst. Just because ATI got 3 cards in the top 5 it doesn't mean that there aren't any others present, in much lower places.

It goes to show though that increasing the quality control for various vendors is a necessity, especially for NVIDIA.

 

You really need to read the whole article to be able to judge the graphs, I had a glance at it and it looks interesting but have not had time to read the whole thing. One of the issues is they are checking rise and fall time and parhelia seems quite sine-like in its graph while the ideal result would be a ractangular wave patter, so the more rounded the curves the worse (checks how good the card is at high contrast areas for example text on white background, if the curve raises slowly you get blurry edges which is bad) another thing is they plot seperate curves for reg, green and blue and if they are not equal you have a shift/unbalance which is obviously bad. They also look at the minum voltage levels that have to be reached on the signaling (this is what causes the really low scores for some cheap budget cards).

Anyway looks interesting but have to read whole article to understand all the issues.

K-

 

The point is, that Parhelia performs worse than G550 - according to tecchannels tests - and that the result for Parhelia doesn´t compare favourably to the Matrox whitepapers on the subject. It would seem that Parhelia doesn´t deliver as promised (in this respect).

3dcgi: you´re too lazy to read it yourself and ask someone else to do it for you? Oh man.

 

You still have to read the article to see why Parhelia or any other high end card may get lower scores in that test. Else explain the rather unfortunate results of the Hercules 9700PRO board compared to it's 8500AIW brother for example.

At the end of the day though -and the author makes it very clear- if you don't run your desktop in ultra high resolutions or don't have a high quality monitor the differences are hard to notice. In a scoregraph the difference between 8.2 and 7.2 in scores might sound "huge", but I doubt it translates into that big real life differences.

 

Hmm, funny point.
What kind of users are the target of Parhelia, would you say?

 

The kind of users that need an at least 7.0+ score for their needs. It's nice that you did single out that very sentence, but there's another that follows it and draws a different overall picture in that paragraph.

Care to elaborate how and to which degree you'd notice which differences between tecchannels scores or more specifically between 7.0 and 8.0?

Rise and fall time is better for the G550, while Amplitude is higher on average for the Parhelia. The RGB curves look equally fine on both accelerators by the way.

 

My point is that Parhelia doesn´t deliver on it´s promises.
If you read Matrox´s PDF on "UltraSharp Output Technology" you will see that they, using partly the same methods as tecchannel, promise better output than the competition (Radeon 8500 and Geforce4Ti), and this already at 1280x1024 since this is the resolution at which they make their tests - just as tecchannel. Whether this translates to tecchannel scores of "7.0+" could therefor be subject for debate, but that is nothing I will lose any sleep over. To me the interesting part is, that Parhelia subperforms vs their PR, according to tecchannels tests.
In this PDF you can also see that they claim rise and fall times for Parhelia very similar to the tecchannel results for the G550. Tecchannel shows this isn´t the case.

Since 2D output quality is more or less the hallmark of Matrox, I´d say this is of interest.

edit: BTW, it would be interesting to see if the results are the same for any resolution with proper aspect ratio.

 

There have been tests at various sites by now.
So far, those I've seen has had the parhelia outperform the other cards (whether this actually translates into a better image is another discussion, and probably strongly monitor dependent).

Does the claims of ONE site with ONE card invalidate the rest of the available data?

Entropy