Page 3 of 7 FirstFirst 12345 ... LastLast
Results 21 to 30 of 66

Thread: 50% slower than AMD! Intel FAIL again with the HD4000 graphic hardware.

  1. #21
    Join Date
    Nov 2008
    Location
    Germany
    Posts
    5,411

    Default

    Quote Originally Posted by scottishduck View Post
    Sounds like someone was trolled into buying bulldozer and needs to justify his awful purchase.
    i do not have buyed a bulldozer yet. and you can read in this forum my favor "bulldozer" i like is the Opteron 6204 with all "bulldozer like" features turned off.

    i wait. in mid 2012 ->2013 amd will push out a quatchannel desktop socket this means i use my "PhenomII x4 B50" and next year i will buy a new system.

  2. #22
    Join Date
    Aug 2007
    Posts
    6,610

    Default

    Maybe you should ask AMD why you can not use a Bulldozer or whatever cpu without external gfx card when you want to use AMD-Vi. All supported chipsets like 890FX, all series 9 do not have onboard vga. And those funny FM1 boards use cpus which are not faster than the normal Athlons (just with much higher price) combined with a gfx core. Of course for laptops those cpus should be enough, at least when you use win to get all accelleration features

  3. #23
    Join Date
    Nov 2008
    Location
    Germany
    Posts
    5,411

    Default

    Quote Originally Posted by Kano View Post
    Maybe you should ask AMD why you can not use a Bulldozer or whatever cpu without external gfx card when you want to use AMD-Vi. All supported chipsets like 890FX, all series 9 do not have onboard vga. And those funny FM1 boards use cpus which are not faster than the normal Athlons (just with much higher price) combined with a gfx core. Of course for laptops those cpus should be enough, at least when you use win to get all accelleration features
    i google it.. and i found nothing like amd-vi?

    do you mean IOMMU ?
    or AMD-V ?

    IOMMU is a chipset feature not a CPU feature, the 890FX supports it and all 900 chips support it.
    AMD-V is the visualisation feature all 45nm and better cpus support it.

    i read about problems with the IOMMU feature in the past and the source is some mainboard builder company disable it and only enable it in the high-priced version.

    but amd is not the source of the problem.

  4. #24
    Join Date
    Apr 2010
    Posts
    1,946

    Default

    Quote Originally Posted by Qaridarium View Post
    i wait. in mid 2012 ->2013 amd will push out a quatchannel desktop socket this means i use my "PhenomII x4 B50" and next year i will buy a new system.
    Sandybridge ex does exactly this.
    I suspect it is one of things that bottlenecks bulli.

    Second is huge intercache latencies and weak internal associativity.

    Third is weak overall power management. The CPU seems to do it right, most of times; but chipset recklessly burns all energy away.

    Fourth and most important absent logic: modify internal CPU scheduler to overload modules with tasks till they are 100% and only then go to next modules.
    That way part of modules will perform 20-100% and rest sleep at 0% till they are really needed.
    This needs internal CPU scheduler to break whatever OS scheduler core-thread association thinks is right and feeds and do it on its own instead - reassign each task manually.
    Last edited by crazycheese; 01-22-2012 at 09:08 AM.

  5. #25
    Join Date
    Nov 2008
    Location
    Germany
    Posts
    5,411

    Default

    Quote Originally Posted by crazycheese View Post
    Sandybridge ex does exactly this.
    I suspect it is one of things that bottlenecks bulli.

    Second is huge intercache latencies and weak internal associativity.

    Third is weak overall power management. The CPU seems to do it right, most of times; but chipset recklessly burns all energy away.

    Fourth and most important absent logic: modify internal CPU scheduler to overload modules with tasks till they are 100% and only then go to next modules.
    That way part of modules will perform 20-100% and rest sleep at 0% till they are really needed.
    This needs internal CPU scheduler to break whatever OS scheduler core-thread association thinks is right and feeds and do it on its own instead - reassign each task manually.
    most of your point are only true on: WINDOWS;;;

    whatever it doesn’t matter in 2-3 months amd will bring a new quatchannel desktop socket and a octachannel server socket.

    the bulldozer successor will perform well on the new sockets with new chip-sets.

  6. #26
    Join Date
    Oct 2007
    Location
    Under the bridge
    Posts
    2,128

    Default

    Q does have a point in this thread. A CPU with a good integrated GPU is *much* better than dealing with Optimus crap.

    AMD's Fusion GPUs are good enough that you don't need a dedicated GPU unless you are a hardcore gamer. Intel's Sandy Bridge may be 10-30% faster than AMD's Llano clock for clock, but Llano's GPU is ~100-200% faster than SB's which makes for a much better all-around performer (that can run modern games in 'medium' settings).

    That's one part of the equation. The other is drivers, where Intel faces significant problems. One year later and Sandy Bridge still fails to vsync properly and still suffers from graphics corruption (esp. in Gnome Shell). The new Atoms are equipped with a DX10/GL3 PowerVR GPU - but guess what. Intel delayed them due to 'driver issues' and finally announced they will only support DX9 (no OpenGL!) and 32-bit Windows. No Linux, no 64-bit, just 32-bit Windows.

    And that's why you don't buy Intel GPUs: their drivers suck. At least with AMD you know you'll get decent support on Linux (with open-source drivers - fglrx sucks) and great support on Windows. No such guarantees with Intel.
    Last edited by BlackStar; 01-23-2012 at 03:09 AM.

  7. #27
    Join Date
    Aug 2009
    Location
    Russe, Bulgaria
    Posts
    504

    Default

    Q, I am waiting for the AMD announcement today

  8. #28
    Join Date
    Aug 2007
    Posts
    6,610

    Default

    So you really call AMD win drivers great? Maybe for DX games, but not for OpenGL. Rage was released 4th oct 11, but there are still only "preview" drivers to run that game. Those drivers with additional char behind usually contain code of the driver that will be out +1 month later. They do absolutely not manage to fix the OpenGL part, no they introduce new bugs, at least for hd 5670 and 11-12. Intel is not perfect too, hopefully there will be a working X stack combination soon. xvba in fglrx is definitely not that advanced, missing h264 l5.1 support. intel already exposes h264 encoding profiles via vaapi, but i am waiting for the killer app for linux yet to use it.

  9. #29
    Join Date
    Apr 2010
    Posts
    16

    Default

    I suggest to Phoronix that they make a new phoronix Platinum Edition, a little more expensive than Phoronix Premium, but with no Qaridarium posts at all.
    I bet they would get more income.

    Of course, they would need to remove the ignore feature, which i'm about to start using now.

  10. #30
    Join Date
    Sep 2007
    Posts
    997

    Default

    Quote Originally Posted by RealNC View Post
    I stopped reading Q's posts long ago. Some months back, he was making some semi-good posts. Now he's just trolling all over the place.
    Good idea.

    One gets the impression that AMD is struggling just from reading comments on various forums of related discussions and the most recent feedback on bulldozer.

    But, even the Quanta report should be reason for concern.

    http://www.theverge.com/2012/1/5/268...e-chip-lawsuit
    http://www.bloomberg.com/news/2012-0...computers.html

    My impression was that AMD OSS drivers was still a mess in Linux compared to using Intel. At least, for using more recent AMD laptops. I notice more reports of problems in the AMD OSS section here than Intel's.

    Apple also uses Intel for their machines.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •