I understand that the bandwith of Matrox Parhelia isn't as good as others but it dose have great expetancy... Now if they produce a Matrox Parhelia 8x AGP version... I would be very happy to see that card and use it...
Announcement
Collapse
No announcement yet.
ATI 9500 non pro 128 v. Parhelia 512 - 128
Collapse
X
-
Originally posted by treilanin
I am utterly confused? ATI cheating in their drivers reducing image quality.... and NVidia doing so much more and better with their Det50s? Somehow we must have this backwards.... ATI has in all honesty done a good job of bringing up the quality of their drivers and Nvidia, regardless of what they tell have had the quality of their drivers plummet with the release of GF-FX series of cards.
To be honest, how can translating code paths in the driver to so called "efficient" codepaths really be a boon for performance. Taking that approach is almost akin to what emulators do, and no matter what anyone says it is always a hit to performance if you are doing more work then you really need to.
Some games are optimized for one card or the other. In my opinion if the image quality is the same / better who cares how it is rendered.
Its like a sorting algorithm. You can sort a 500000 record with many different algorithms. Using a selection sort is a lot slower than using a merge sort, but merge sort in general is much faster, and their outputs are still the same. I'd choose merge sort over selection sort, because its more efficient.
For code path, I think they are the same. Its like taking advantage of the hardware for both quality and speed. Its hard to translate one code path to another, but the enhanched code path is better without sarcifing quality, so who cares.
The whole problem in here is standardization. There is no standard way of rendering. THe program maybe better for one graphics card than the other, because the benchmark itself may use a code path designed for one of them over the other. If you are benchmarking with such program, that's really not fair. When they design the program its better for one card than another, so such a program is no good to compare the two cards anyways. I think translating codepath to optimize how stuff is render is a good thing.
Ultimatly what it matters is the output, not how it is processed. Skipping some stuff in ur picture is not good optimization is not good, because it affects the output. But not enhanching the way stuff is better for greater speed and better/same quality output, that's just stupid not to. Better to be smart than to use brute force.
Comment
-
If you play C&C Generals a lot, don't get a 9600 (Pro) they seem to have Hardware Cursor issues with some games (such as C&C Generals), which means you have to disable hardware cursor and get a stupid white flickering cursor instead.Main Machine: Intel Q6600@3.33, Abit IP-35 E, 4 x Geil 2048MB PC2-6400-CL4, Asus Geforce 8800GTS 512MB@700/2100, 150GB WD Raptor, Highpoint RR2640, 3x Seagate LP 1.5TB (RAID5), NEC-3500 DVD+/-R(W), Antec SLK3700BQE case, BeQuiet! DarkPower Pro 530W
Comment
-
Originally posted by iampedro
I understand that the bandwith of Matrox Parhelia isn't as good as others but it dose have great expetancy... Now if they produce a Matrox Parhelia 8x AGP version... I would be very happy to see that card and use it...Last edited by gangster; 13 November 2003, 18:56.P4b@2.7, AOpen ax4spe max II, 4X Parhelia 128 with Zalman zm80c and fan -or- ATI Radeon X800GTO, 1024mb.
Comment
Comment