The AMD oboard graphics are actually alot faster technically speaking than any Intel offerings, even the latest ones. The catalyst driver should definitely be thrown into the mix along side with the open source driver. Even still, catalyst in Linux does not perform even close to the Windows offering (overall) and it really pains me, because they would be the best bang-for-the buck systems otherwise.
Announcement
Collapse
No announcement yet.
Intel Haswell HD Graphics 4600 Performance On Ubuntu Linux
Collapse
X
-
Originally posted by mannerov View PostI don't understand your point. I don't think it's insulting wanting to have an other line showing performance with the catalyst driver, since it will give a not unbiaised comparison (in all other benchmarks, A10 performs better than hd 4000).
Before seeing it was not the catalyst driver, I thought there was an issue somewhere.
A much larger comparison of Haswell graphics against various discrete AMD Radeon and NVIDIA GeForce graphics cards will come in a later Phoronix article where testing will happen on the latest open-source and closed-source graphics drivers.
Comment
-
Originally posted by enfocomp View PostEven still, catalyst in Linux does not perform even close to the Windows offering (overall) and it really pains me, because they would be the best bang-for-the buck systems otherwise.
I need to upgrade my PC, and on the Hardware side, I see much better value in getting an A10-6800k (rather than an i5 Haswell), but the software is most likely going to kill it. I have nothing too philosophical against using binary drivers, but I've had so many little issues with both Nvidia and ATI that I avoid them like the plague. So, as it stands, the Haswell costs me 50% more and gives me twice the performance in graphics, and 50% more CPU. So, it's kinda like a better value. Of course, the total cost is higher, so I'll need to figure out whether I need the extra performance, or I can live with the A10-6800k. Decisions, decisions.
Comment
-
Originally posted by mendieta View PostI have nothing too philosophical against using binary drivers, but I've had so many little issues with both Nvidia and ATI that I avoid them like the plague. So, as it stands, the Haswell costs me 50% more and gives me twice the performance in graphics, and 50% more CPU. So, it's kinda like a better value. Of course, the total cost is higher, so I'll need to figure out whether I need the extra performance, or I can live with the A10-6800k. Decisions, decisions.
Comment
-
Originally posted by shmerl View PostIs Intel going to put Iris Pro chips in their desktop CPUs? I'm hesitant to buy the Haswell CPUs when their integrated GPUs aren't the best. But if they'll never do it - no point to wait may be.
Comment
-
Originally posted by oleid View PostThis last-gen AMD APU should perform faster when using the latest improvements as the hand-tuned shaders...
Comment
-
OTOH, AMD is using a chip similar to the A10-5800k, but with 8 CPU cores, for the new Playstation. Why the F*** are they not releasing that to the public? That would be a no brainer for a fantastic value, for an all around balanced APU for general computing and graphics. If they can sell it for ~ $200, they'd have a slam dunk. I guess power issues must be the impediment, but hey, they did it for Sony!
Comment
-
Originally posted by mannerov View PostWhy does the A10-5800K performs so poorly?
I checked some directx benchmarks on internet and saw that it should perform better than the hd4000 and the hd4600.
EDIT:
I found why: the benchmark here is done with the open source driver.
I think adding a line for the A10 and the catalyst driver would have been appropriate.
It performs bad because, number one, Michael needs to either tune ondemand governor, or force performance governor! This may boost up to 50% performance in some tasks.
Number two, the current driver setting is to always fallback to safe clocking, because there is no temperature-aware powermanagement in opensource (yet). If one allows the clock go unlimited(safe if one has good cooling), the performance is doubled. Check this out.
Originally posted by mendieta View PostYou hit the nail on the head. I think Intel's gap Windows vs Linux (OS) is smaller than the gap in AMD (Windows vs Linux Catalyst). On top of that, you have the huge gap (a factor 2 to 3 in many cases) between AMD Catalyst vs AMD OS. So, yes, whenever that comparison is published, it will be very useful. Haswell in Linux should work much faster as they add proper optimizations.
1) Linux catalyst often performs faster than windows catalyst. But it does have less features!
2) Linux OS is 100%-70% of Linux catalyst. "factor 2 to 3" was a year ago, before Vadim's patches, before ondemand bottleneck discovered, with older mesa library, before driver optimizations from Marek.
The cutting edge Intel has over AMD on opensource font is working powermanagement and very good development rate.
So, unless AMD releases dynpm for opensource drivers (they released UVD, so its possible!), for those who are tired of Catalyst!, Intel would be not-so-cheap, but definitely worthy pick!Last edited by brosis; 06 June 2013, 04:27 PM.
Comment
-
Originally posted by enfocomp View PostThe AMD oboard graphics are actually alot faster technically speaking than any Intel offerings, even the latest ones. The catalyst driver should definitely be thrown into the mix along side with the open source driver. Even still, catalyst in Linux does not perform even close to the Windows offering (overall) and it really pains me, because they would be the best bang-for-the buck systems otherwise.
Comment
Comment