Jump to content
HWBOT Community Forums

bolc

Members
 View Profile  See their activity

  • Posts

    689

  • Joined

  • Last visited

  • Days Won

    3

 Content Type 

Posts posted by bolc

    DDR1 overclocking on DFI Lanparty Nfroce4 ultra - how is ram = 1.2 x FSB possible ?

    in General overclocking

    Posted

    Hi all,

     

    I am not spotting anyone in particular, but I wonder, how is it possible to get the x 1.2 ram multiplier...?

    I though the max was 1:1, hence ram = fsb frequency. but currently there are lots of subs with ram = 1.2 x fsb.

    I have DFI lanparty nforce4 ultra and nowhere to find this beautiful x1.2 coefficient ...

     

    https://hwbot.org/submission/2646636_kotori_memory_frequency_ddr_sd_ram_447.8_mhz

    https://hwbot.org/submission/2308828_christian_ney_memory_frequency_ddr_sd_ram_441.6_mhz

    https://hwbot.org/submission/2308808_i.nfrar.ed_memory_frequency_ddr_sd_ram_440.5_mhz

     

    what is the secret...?

    ++

    bolc - Celeron G470 @ 2143MHz - 0.6 fps HWBOT x265 Benchmark - 4k

    in Result Discussions

    Posted

    If my imc would behave at 2330 mhz+ I could run bclk as high as 109.6 as I do in my 2.51 fps run and do .61 fps. Truemonkey is running on the older version of x265 that scores worse. Plus there are a few tweaks that he's missing as well as georgeOC from the look of it ;) At the very least I'm on my 3rd chip, I still have another to test with 3 more in the mail and a dice pot for the mem coming I'm fairly confident I can set .61 or .62 fps.

     

     

    don't do that :D :cry:

    Boraburner - Core i5 3317U @ 3342MHz - 3342 mhz CPU Frequency

    in Result Discussions

    Posted

    Salut et welcome chez HFR :)

     

    J ai peur que ta soumission ne soit pas valide, un screen hwinfo n a pas de valeur, et la validation cpu-z indique 2400 mhz. Si tu désactive 1 core (n active qu un seul core) tu seras sur le coeff x36 donc 2600 Mhz possiblement validable. Pr cela, tu ouvres msconfig.exe et dans les options avancées du boot, choisis numproc =1.

    ou alors d sle biops, si tu as l option, tu desactives le multicore.

     

    Je te conseille de retirer ta soumission, ui va etre supprimée tot ou tard.

    Cdlt, Bolc

     

     

    yosarianilives - Celeron G470 @ 2166MHz - 2.49 fps HWBOT x265 Benchmark - 1080p

    in Result Discussions

    Posted

    I believe it's the rounded, as it is now if I pushed ram even harder I could take a few seconds off and technically beat him however you're right I have to take off like 30s to actually beat him. I have the advantage of a better x265 and psc that's surprisingly decent for sandy so if I could do 109 bclk I'd probably have it. Ordered 2 more chips to bin after setting this :)

     

    109 would be nice :D 2 chips my way today :D

    yosarianilives - Celeron G470 @ 2166MHz - 2.49 fps HWBOT x265 Benchmark - 1080p

    in Result Discussions

    Posted

    Well done !

    Do you know if it is the rounded FPS that counts for ranking, or the real computing time? cause you will never exceed 0.6 fps (you'd have to go above 0.605 to get 0.61 rounding and there is some a long way to go with g470 to get there.... But at 0.600 real fps you d be first since True Monkey s score

    http://hwbot.org/submission/3451078_true_monkey_hwbot_x265_benchmark___4k_celeron_g470_0.6_fps

    is actually 0.5996 fps for 47'10 sec (no ms on the 2.0 version...)

    lukisek20 - Xeon E5450 4000 Mhz - 9s 29ms Superpi 1M

    in Result Discussions

    Posted · Edited by bolc

    Here are submissions by Lukisek20 for Xeon E5450:

     

    http://hwbot.org/submission/3758065_lukisek20_pifast_xeon_e5450_15sec_740ms

    http://hwbot.org/submission/3759127_lukisek20_superpi___1m_xeon_e5450_9sec_29ms

    http://hwbot.org/submission/3758074_lukisek20_superpi___32m_xeon_e5450_10min_0sec_741ms

    http://hwbot.org/submission/3758062_lukisek20_wprime___32m_xeon_e5450_8sec_631ms

    http://hwbot.org/submission/3758063_lukisek20_wprime___1024m_xeon_e5450_4min_12sec_518ms

    http://hwbot.org/submission/3753187_lukisek20_cinebench___r11.5_xeon_e5450_5.95_points

    http://hwbot.org/submission/3751208_lukisek20_cinebench___r15_xeon_e5450_356_cb

     

    to me it seems only the CBR15 (last one) is in line with other results, all others seem far too good to be true.

     

    For instance compare with TAGG superpi1M at 4800+ Mhz for 9.5 sec vs 9.029 for lukisek20 @ 4000 Mhz...

    http://hwbot.org/submission/3580246_tagg_superpi___1m_xeon_e5450_9sec_515ms

    GPUPI - SuperPI on the GPU

    in Benchmark software

    Posted

    IIRC HD 4800 stopped working with GPUPI long time ago. I think when version 2.3.x was released.

     

    @_mat_

    could you please also make 32-bit version of GPUPI 3.1 if it becomes mandatory? There is quite a lot of 32-bit only NetBurst CPUs otherwise capable of running GPUPI. And it is always nice to get free points and gold cups :D

     

     

    trying on 2.2 gave me no device supported, while it used ot. I will look into the driver issue as far as I am concerned with catalyst 11.12 and olders...

    GPUPI - SuperPI on the GPU

    in Benchmark software

    Posted · Edited by bolc

    If the device is not shown, it is a driver issue. The enumeration of devices is a pretty basic thing an relies on what the OpenCL driver returns.

     

    Just to be sure: Have you tried Luxmark or other OpenCL applications? Do you have the same problems there?

     

    Hi

    Thanks.

    the strange thing is that some suceeded to detect the card

    http://hwbot.org/submission/3319863_nicegab_gpupi___1b_radeon_hd_4850_gddr3_14min_6sec_376ms

    with catalyst 13.9, opencl 1.2

     

    on my hd7950, installing catalyst 15.7.1 installs opencl driver.

     

     

    it seems this carsd is only compliant with opencl 1.0 (these sites say 1.0 but gpupi worked in 1.2 so I guess 1.x is ok....?)

    https://www.techpowerup.com/gpudb/493/radeon-hd-4850

    https://community.amd.com/thread/152535

    here they dismiss 48xx from 1.1, suggesting that 1.0 is the only driver good for 4850.

    https://www.mql5.com/en/articles/690

    but since it worked on gpupi for several hwbot subs, 1.2 should be ok.

     

     

    the only way I know so far to install opencl is via catalyst center. it installs OpenCL Driver 10.0.937.2.

     

    perhaps there is another way to force 1.2 installation instead of 2.0 ?

     

     

    THERE MAY BE HOPE HERE:

    http://www.geeks3d.com/20091221/how-to-enable-opencl-support-on-nvidia-and-amd-platforms/

     

    I will try luxmark too.

    Cheers

    GPUPI - SuperPI on the GPU

    in Benchmark software

    Posted · Edited by bolc

    If you want to test as well, feel free to download GPUPI 3.1 right here: https://www.overclockers.at/downloads/GPUPI%203.1.zip

     

     

    Hi :)

    I have tried running Win7 64 with HD4850, catalyst driver 13.9, latest on amd site, with gpupi 3.0.1, and the device is not seen un gpupi. It seels intel opencl and amd opencl for the cpu only.

    same thing with 2.2, 2.3.4 legacy and 3.1 :o

    any hints where to look after?

    Thank you

    Cheers

    ProKoN - Core i5 3470 @ 4159.8MHz - 858 marks XTU

    in Result Discussions

    Posted · Edited by bolc

    Only thing I can think of is if you aren't setting ALL timings manually it trains looser secondaries at cl7.

     

     

    I fix them all manually. trrd 4 trtp 4 twtr 7 twr6 trfc 86 tckl 4 twcl8 faw 18 for secondary

     

    mrc fast boot is enabled on my board i think, can put it off to see if...

     

    nonetheless, subtimings can make 50+ pts difference ...?

    Xeon L3014

    in Sold

    Posted · Edited by bolc

    u r crazzyyyy.... !!!

     

    beware because 2019 is faaaaaaaaaaaaaaaarrrrrrrrrrrrr from now....

     

    and you won t have enough time in 2019 to bench all this :D

    send me one L3014, and I will send it back to you in working condition ! ;):D

     

    PS: curious what those Trident ? = 3 kits of G.skill 3x2 2000 cl9 will give, if they are psc or elpida (serial numbers 4-7 or 5-8 are 104x or 064x/065x/066x)

    My favorite and best ram are g.skill, and trident / tridentx, forever...!

    Luumi - Core i5 3470 @ 4220MHz - 853 marks XTU

    in Result Discussions

    Posted

    40x is available for 1-2 cores only, with 3x it is 39x and with 4c 38x

     

    to enable 1-2-3c under lighter loads, enable C3 c-state option (that one is enough, leave other c-states disabled)

     

    hence during the prime workload of xtu, the cpu will be 38x but in between it will go to 39-40x, perhaps making intermezzos a little faster ; on the overall score XTU it changes nothing imo...

    ProKoN - Core i5 3470 @ 4159.8MHz - 858 marks XTU

    in Result Discussions

    Posted

    Yes sure.

     

    One thing I am heavily suprised about, and I can/will substantiate my thoughts with screenshots if you want, is that benching with CAS7 always gives much much worse scores for me when using XTU !!!

     

    For instance, the 3470 benched at 107.8-108ish bclk, and 2200+ CAS7 = 7-10-6-24-1T trfc=86 and fairly tight other subs, will barely give 798-800pts, while the same frequency on CAS8 (8-10-8-24-1T or 8-10-8-26-1T) will make an easy 850+ pts on 2133 or 2200 ram multiplier.

     

     

    If I bench with Prokon (2185 Mhz @ 7-10-6-24-1T) or with Luumi's settings (2300 Mhz @ 7-10-6-28-1T http://hwbot.org/submission/3680063_luumi_xtu_core_i5_3470_853_marks), I will never exceed 800 pts, no matter if I use the 2000 Mhz (1:10) ram multiplier like Prokon does, or the 2133 Mhz 1:8 multi like Luumi does...

     

    The same happens on Haswell. I tried Win7 64 or Win7 32 (btw no difference between both... and no difference between 8.1 32 or 10 32 and 7 64... superfetch + maxmem applied)

     

    Any thoughts why they can, and why I can't?

     

    Ps: My ram can handle 2300+ cas7 (g.skill pi 2200cas7) and 2600+ cas8

    Xeon L3014

    in Sold

    Posted · Edited by bolc

    You will if you want that one. The price in that link is about the going rate.

    ;)

     

     

    I changed my mind in the meantime... :)

    and wait for l0ud_sil3nc3l / The Purge to tell me if he has any...

×
×
  • Create New...