What Shall We Benchmark Next? Let Us Know!

[Michael]

And yes, Wine is a very real and good possibility especially as the Windows PTS support evolves there will be new Wine-compatible test profiles. It should be possible to track Wine as well on a per-commit basis.

[darkbasic]

It's not a job for Phoromatic, but I'dd like to see a benchmark of linux-2.6.33 compiled with gcc vs icc (linuxdna).

[jakubo]

i guess anything that might help in finding regressions in any way. but it ll be hard to find anything that evolves as fast as the kernel, so that daily snapshots would make sense.
what im missing personally is to see other hardware drivers being benchmarked, like WLAN cards, or chipsets. i would like to see if there are differences in reliability and performance, and if theres any development.
maybe different programs: wicd vs. network manager
that would be rather ideas for testing once in a while.
But a transfer rate/data loss/signal strengh tests for instance would be interesting.
i have the feeling its being neglected a bit.

[MuPuF]

Wine is a really good idea.

But, as fedora raw-hide is not stable enough, you can use archlinux (without the testing repo) and you should be able to monitor the performance of an entirely up to date system (latest stable version of every program).

[Michael]

So you could then test with lower spec cards from AMD and NVidia, and whatever graphics Intel have, and see what it looks like with the native performance, and the wine performance in the drivers with Direct3D and openGL.

Unlikely to do all of that for a tracker unless there is significant upstream interest in the tracker and also an increase in use of the affiliate shopping links / Premium subscribers / anything to justify the electricity for a number of systems dedicated to that task.

[cruiseoveride]

We are seeking requests for trackers to continuously monitor performance, not individual performance article benchmarks. Post those requests in another thread as I will just be ignoring them in this thread.

Wow, really?
Michael--;


WineD3D++;

[DoDoENT]

I vote for Wine and for Radeon/RadeonHD.

It would be interesting to see the progress of open source radeon drivers on daily basis.

And what about making an agreement with AMD to provide you with daily builds of their fglrx driver? If they agree, you could test their daily fglrx build on various AMD cards (HD 2xxx, 3xxx, 4xxx and 5xxx) - it may speed up the development of AMD drivers. They may even pay your electricity bill (or at least part of it).

Maybe the similar agreement with NVIDIA?

[duby229]

I think I'm gonna have to vote for wine as well. I would also like to put another vote in for tracking one of the *BSDs

Wine usually has such a huge number of regressions and fixes during each cycle I think that tracking it with Phoromatic would be an excellent idea. Tracking one of the BSDs I think would have less imact for me then wine, but it would still be a great way to see valid repeatable comparisons between the available kernels. Another cool thing is at some point in the future other kernels.

[Michael]

And what about making an agreement with AMD to provide you with daily builds of their fglrx driver? If they agree, you could test their daily fglrx build on various AMD cards (HD 2xxx, 3xxx, 4xxx and 5xxx) - it may speed up the development of AMD drivers. They may even pay your electricity bill (or at least part of it).

Maybe the similar agreement with NVIDIA?

All the major IHVs have their own internal test farms already, many of which use the Phoronix Test Suite as part of their process.

[Michael]

Wow, really?
Michael--;

They are two entirely separate and distinct products, I don't see what the fuss is about with keeping a thread on-topic.