Results 1 to 19 of 19

Thread: Intel 965 (P5BDLX) vs Intel 975(P5W64-prof) *User review*

  1. #1
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247

    Intel 965 (P5BDLX) vs Intel 975(P5W64-prof) *User review*

    Chipset Intel 965 VS 975 with E6400 by Giorgio Benetti alias Benna65 & Vincenzo Parrello alias v_parrello


    www.amdclockers.it/forum


    I apoligize for my bad english... i just translate this review that has been made from one of mine moderator (vparrello) and another user (Benna65).
    I have to thanks both them, cause they did an awesome job!

    Intro

    The following thread is made to compare this two asus mainboard, equipped with the intel 965 (P5B Deluxe) and with the intel 975 chipset (P5W64 Professional).
    Both mainboards are the High end by ASUS!
    As we all now, this two chipset run very different in overclock when wh pull up the FSB. And in particular mode, the 965 mainboards let us to reach very high fsb when they change the strap from 1066 to 1333 at the 401MHz wall.
    In the other hand, the 975 mainboards usually wont reach the same fsb score as the 965 ones, but they keep the same aggressive chipset timings setup, and the strap is the same when we go over the 401MHz wall.
    What we want to show in this review, is to see how do these two different chipsets with a particolar look at their performance growin up the fsb. We want to answer the usual question of everyone who doens't know if it is worth or not to buy a 965 mainboard instead of a 975!



    Configuration settings:

    We made these tests on two different systems. It doens'nt only change the mainboard, but the memories too (but they have the same D9 chip) and the HDD too, but they are a lot similar in all the other parts. It is normal that it will not be a perfect review, cause there will be some error for sure, but the meaning of this review is to find a overclock performance trendlines to compare. We make different test at different fsb!




    Here it is the two system we used:



    What do we have to see in the scores:

    Here it is the scores we've got at various frequencies with different benchmarking tools:





    And here it is the trendlines we talked abt before, now we can see the real difference b/w the 965chipset and the 975:













    From the graphs we found two important points:

    1. About the bandwith performance and the latency performance the 975 chipset beat the 965 everywhere;

    2. about the calculum performance the intel 965 chipset beat the 975 till the 400MHz wall, then the 975 keeps to get better and as we all know on the 965 starts the 1333 strap and it looses performance. The only test for the calculum performance where the intel 975 beats the iontel 965 is for the 1M super-pi, and it is because this test it isn't long enough, and we think the "problem" is in the bandwith and in the latency performance. While in the other tests is true what we said before.

    Conclusions

    At this point the conclusions could be differents because of what we read in the scores. We have to say that from this test doens't come out a real winner, but two different platform to choose from keepin in mind this points:
    • price difference to spend if we want a 975 mb or a 965mb ;

    • the frequency we want to make our cpu to work (below the 400MHz wall the 965is better, above the 975 is the best) ;

    • the better functions of the 975 chipset (i.e. crossfire ready);

    • what kinda cpu we are gonna buy or we have, the max multiplier and the max fsb reachable from it.


    Test Screenshots

    FSB 266 chipset 965, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra



    FSB 266 chipset 975, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra

    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  2. #2
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    FSB 300 chipset 965, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra



    FSB 300 chipset 975, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra



    FSB 350 chipset 965, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra



    FSB 350 chipset 975, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra

    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  3. #3
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    FSB 470 chipset 965, SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra



    FSB 470 chipset 975, in ordine SuperPI, CPUBench2003 e Everest, ScienceMark 2.0, Sandra

    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  4. #4
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    reserved 2
    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  5. #5
    Xtreme Addict
    Join Date
    Sep 2006
    Location
    Surat, India.
    Posts
    1,309
    Very Very Nice Results keep then coming.
    Sound: Asus Essense ST | Wharfedale Diamond 9.1 | Norge 2060 Stereo amp | Wharfedale SW150 sub (coming soon)
    Camera Gear: Canon 6D | Canon 500D | Canon 17-40L | Canon 24-105L | Canon 50mm f1.4 | Canon 85mm f1.8 | Rokinon 14mm f2.8 | Sigma 10-20EX HSM | Benro A3580F + Vanguard SBH250 | Bag full of filters and stuff

  6. #6
    Xtreme Mentor
    Join Date
    Aug 2005
    Location
    Boston, MA, USA
    Posts
    3,000
    Can you test the 1066 strap versus the 1333 strap?

  7. #7
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    Quote Originally Posted by uOpt
    Can you test the 1066 strap versus the 1333 strap?
    that is what we did man...

    Marco
    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  8. #8
    Xtreme Enthusiast
    Join Date
    Apr 2005
    Location
    Germany, Bremen
    Posts
    541
    nice test and now i like my board more than before

  9. #9
    Xtreme Enthusiast
    Join Date
    May 2004
    Location
    San Diego, CA
    Posts
    761
    So can someone break it down into simple terms?

    Like if you want a 6600 and 3GHz use this MB, if you want a 6400 and want 3GHz use that MB.

    I want a 6600 at 3GHz and need to know which MB would be best for me.

    intel i7 3770k 4.40GHz
    ASUS Maximus V Extreme
    EVGA GTX 680 FTW
    16GB GSKill 2400MHz
    (1) OCZ Vertex 3 60GB SSD
    (1) Samsung 840 Pro 256GB SSD
    (1) 1.5 TB WD Caviar Green SATA
    Sound Blaster X-Fi Titanium Fatal1ty Professional
    Corsair 700D
    Corsair AX850
    Water Cooled (EK blocks on CPU and GPU, TA 120.4, DDC-1+ with XSPC ResTop)

  10. #10
    I am Xtreme
    Join Date
    Dec 2002
    Posts
    5,942
    terrific writeup, you forgot to mention that if you want over 450fsb the 965 may be better because it will go 500+ alot easier.

    too bad we cant control the straps on the 965 i have a feeling it would really rock if it did.

  11. #11
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    6600 or higher multiplier => chipset 975
    6400 or 6300 => chipset 965

    In any case if you want to save some money go with the 965 that you won't loose performace till 400Mhz Fsb

    Marco
    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  12. #12
    I am Xtreme
    Join Date
    Aug 2003
    Location
    Cherry Hill, NJ
    Posts
    8,296
    Great review principino More proof that 965 is slow
    Phenom II 940 BE / ASUS M4A79 / HD5770 Crossfire
    3770mhz CPU 2600mhz NB | DDR1040 5-5-5-15 | 900/1250

  13. #13
    Xtreme Member
    Join Date
    Apr 2005
    Posts
    111
    Wow very good information. Thank you

    This will help others.

  14. #14
    Xtreme Member
    Join Date
    Jun 2006
    Location
    Belgium - Brussel
    Posts
    237
    Good review

  15. #15
    Xtreme Member
    Join Date
    Jan 2006
    Location
    Montreal
    Posts
    376
    THanks for this review principino1984
    |::| i7 920 3849A914
    |::|EVGA Classified 760
    |::|G.Skill Trident 6GB DDR3-2000
    |::|BFG GTX 275
    |::|G.Skill Falcon 2 128GB
    |::|WD BLACK 1TBx2
    |::|Ultra X3 1000W
    |::|Powered by Heatkiller LT/Ek-Classified/EK 275GTX /Feser x360/MCP355x2
    |::|LCD: Samsung T260


    "Sharing your knowledge is always the best way to learn new things". OPB

  16. #16
    Corsair Components, Inc.
    Join Date
    Dec 2003
    Location
    San Francisco, CA
    Posts
    5,066
    Nice write-up, principino1984. It's interesting to see, that 965P produces better 32M times below 400MHz than 975x... Hmmm... I also noticed that Everest bandwidth and latency results are much better on 975x than 965P...

    It would be worth pointing out, that P5B DLX's drop in performance above 400MHz FSB can be bypassed by booting the system @ 400MHz and using ClockGen to up FSB. Most unmodded P5B DLX (with some good effort on user's side) would reach 450MHz (1:1) or around 420-430MHz (4:5) or 420+Mhz (2:3, limited by memory) using this method and therefore retaining benefits of tighter NB latencies and keeping up with 975x. Then vmch-modded P5B DLX could be even capable of hitting 470MHz FSB (1:1) or a bit higher.
    Again, good job man
    Last edited by bachus_anonym; 11-03-2006 at 08:05 AM.
    I work for Corsair Components, Inc.

  17. #17
    Xtreme Member
    Join Date
    Feb 2005
    Location
    Italy
    Posts
    247
    Quote Originally Posted by bachus_anonym
    Nice write-up, principino1984. It's interesting to see, that 965P produces better 32M times below 400MHz than 975x... Hmmm... I also noticed that Everest bandwidth and latency results are much better on 975x than 965P...

    It would be worth pointing out, that P5B DLX's drop in performance above 400MHz FSB can be bypassed by booting the system @ 400MHz and using ClockGen to up FSB. Most unmodded P5B DLX (with some good effort on user's side) would reach 450MHz (1:1) or around 420-430MHz (4:5) or 420+Mhz (2:3, limited by memory) using this method and therefore retaining benefits of tighter NB latencies and keeping up with 975x. Then vmch-modded P5B DLX could be even capable of hitting 470MHz FSB (1:1) or a bit higher.
    Again, good job man

    that is correct man, if you want to make some bench you can boot @399 and then raise up the fsb with clockgen!

    thanks man, I appreciate!

    Marco
    XtremeHardware.it
    Asrock Italian Official Support Forum, Enermax and Sapphire Support Forum

  18. #18
    Xtreme X.I.P.
    Join Date
    May 2004
    Location
    Hammel/Denmark
    Posts
    2,082
    Very interesting review
    The whole world's hip and thats not cool !

    Rip Van Hippy

    Doesn't expecting the unexpected make the unexpected the expected ?

    Light travels faster than sound. This is why some people appear bright until you hear them speak


    -------------------------

  19. #19
    OCTeamDenmark Founder Nosfer@tu's Avatar
    Join Date
    May 2004
    Location
    Denmark, Copenhagen
    Posts
    2,368
    Quote Originally Posted by afireinside
    Great review principino More proof that 965 is slow
    Correct, BUT the 975 SUCK at OC.

    The wall at 400 fsb is so lame

    principino Pure skill to make such a review.
    More of those and you will reach the higest level og skill here on XS
    Former owner of OCTeamDenmark.com
    MSI MOTHERBOARD!!!!!!

    Linkedin


Posting Permissions

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