Shopping cart    |      
Page 1 of 2 12 LastLast
Results 1 to 10 of 15

Thread: A6-3420M vs. A6-3400M

  1. #1
    Join Date
    Feb 2012
    Location
    Sweden
    Posts
    6

    Default A6-3420M vs. A6-3400M

    I am really confused regarding the PerformanceTest benchmark score for AMD A6-3420M. The score for A6-3400M is 3375 but this updated CPU with 100MHz higher clock frequency (both regular and turbo) get a MUCH lower score, currently 1964.

    I am about to buy a new laptop and don't need a "super fast" one, I just want a cool and quiet laptop, preferably with a long battery time. But I would like to know why the score is lower (and a lot at that) on the newer A6-3420M.

    Does anybody have any information/answers that could help ease my mind?

    Zigge

  2. #2
    Join Date
    Oct 2007
    Posts
    234

    Default

    From the link below it looks like the only difference is the 100MHz higher clock frequency so the a6-3420m should be slightly faster. After looking at the baseline files available there are only a few a6-3420m's in the database but tons of the a6-3400m. This may paint a false picture. Its possible because of the small number of results for the a6-3420m all tested laptops were tested in a low power mode so the APU was not running at full speed. Also I found some overclocked a6-3400m's that could be skewing the results. No guaranty but the a6-3400m will perform better after more people submit results. Also if you can wait someone from Passmark will most likely chime in. Could be a database or cpu ID problem also? If so Passmark is fast at fixing either problem if one exists. Oh if you are purchasing from a B&M store run PT on the demo model your looking at buying.
    http://compare-processors.com/amd-a6-3400m-vs-a6-3420m/1885/

    Bill
    Main Box*i7 930*GA X58A-UD3R R2*3x4 gig Patriot DDR3 1600EL*EVGA GTX 460 1 gig*Intel X25-M G2 80 gig*WD Green WD20EARS 2 TB*ASUS DRW-24B3LT*Samsung SH-S223L*Corsair AX750 PSU*Rosewill Challenger case*Windows 7 Ultimate 64 bit SP1

  3. #3
    Join Date
    Jan 2003
    Location
    Sydney Australia
    Posts
    4,140

    Default

    Yes, I think wonderwrench is correct.

    For the AMD A6-3400M CPU we have 507 samples. 94% are 64bit. There is a grouping of results at a CPUMark of 2150, and another grouping at 3500. The probably represents the 2 main power states. Battery power and mains power.

    For the AMD A6-3420M CPU we have only 8 samples. 87% are 64bit. There is a grouping at 2250. But nothing in the 3000+ range.

    I think there are 4 issues.

    1) The limited number of AMD A6-3420M samples were probably all run on battery power with the CPU in a low power state. Can't be 100% sure but this is likely. This effect disappears once we have 100s of samples. This is the major influence I think.

    2) The higher % of 32bit results on the AMD A6-3420M slightly pull down the result. My estimate is only about 3% however.

    3) We aren't correctly detecting excluding all the AMD A6-3400M chips that are overclocked. We made some changes in this area today. But more are still required. My estimate is that this might drop the result of the AMD A6-3400M by about 10%.

    4) The A6-3420M has a high percentage of 'bad' results. Probably badly configured machines, or overheating machines. 2 of the 8 samples were bad. I have deleted these. So when the charts update overnight this will be reflected in the charts. There were a couple of bad ones as well for the A6-3400M, but < 0.5% of the samples.

  4. #4
    Join Date
    Feb 2012
    Location
    Sweden
    Posts
    6

    Default

    Thank you so much for the extensive answer and of course, it makes perfect sense. Perhaps the benchmark program should notify the user that the computer is currently using the battery and the benchmark score will only be accurate when running on the power supply.

    The AMD may not have the fastest CPU:s but the combined CPU/GPU solution is in my opinion the best deal for the money at the moment. And I must say that their AMD Trinity demo @ CES 2012 was truly amazing even though I think the A6-3420M will be sufficient for my needs. It depends on how long I have to wait for the right model to be available, HP ProBook S or Lenovo Thinkpad Edge are the ones I am looking for (only anti-glare screen for me).

    Once again, thanks for putting my mind at ease!

    Zigge

  5. #5
    Join Date
    Jun 2012
    Posts
    2

    Default

    I'm actually glad I googled this. Because I was having the same issue with the same cpu A6-3400M and A6-3420M. This made me a hesitant to buy a laptop,I had to choose between two same laptop which had the same GPU but these two CPU. In conclusion from this thread I should go on with A6-3420M for performance right? or does anyone have any new info on this? the A6-3420M cpu still does show a smaller sample than the A6-3400M but clearly increased.

  6. #6
    Join Date
    Jan 2003
    Location
    Sydney Australia
    Posts
    4,140

    Default

    More samples have been collected since the initial posts above. The scores have moved a closer together but not by as much as I expected.

    I think the results might also be suffering from the consequences of this AMD CPU bug. Maybe a much higher percentage of the the vendors selling the A6-3420M have patched their systems (reducing the performance).

  7. #7
    Join Date
    Jun 2012
    Posts
    2

    Default

    I'll go with the A6-3400M then just to be safe.

  8. #8
    Join Date
    Jul 2012
    Posts
    1

    Default A6-3420m

    The CPU mark for my Asus K53T - A6-3420M with 6520G + 7670M 1GB and 6GB RAM is:
    2297.8

    It has dual graphics but when doing the test it did not use the discrete gpu. Only the integrated was used. High Performance setting was used for the laptop.










    http://www.staples.ca/ENG/Catalog/cat_sku.asp?CatIds=66%2C67,3199&webid=726133&affix edcode=WW
    Last edited by Khmerog; 07-02-2012 at 09:38 PM.

  9. #9
    Join Date
    Jan 2003
    Location
    Sydney Australia
    Posts
    4,140

    Default

    The images are not visible and the link goes nowhere. So I am not sure what you are asking about.

  10. #10
    Join Date
    Feb 2012
    Location
    Sweden
    Posts
    6

    Default

    It's a bit embarrassing but I have to admit it, I still have not purchased my laptop. The passmark scores eventually showed up and in combination with the "poor" A6-3420M results, I decided to wait for Trinity. But this has been quite a stressful choice ... because they have not turned up in laptops at a pace that one would have expected (I know there is always some delay but come on ...). Even though I really like AMD and think they have a very compelling product here, I cannot wait forever.

    The upcoming Windows 8 has some really nice updates but Microsoft have ruined this by forcing their new interface (formerly known as Metro) upon us all and this seems like a disaster in a desktop environment with mouse and keyboard. So there it is, hardware is late and Windows 8 (which I obviously don't want at this point) is getting closer (Oct. 26?).

    I might have to buy Intel if the A8-4500M or A10-4600M doesn't show up in the laptops I am looking for. My requirements: 13.3 or 14 inch with "adequate" performance, good battery life and most importantly, anti-glare screen.

    My list of interesting models so far:

    • Lenovo Thinkpad E335 (I would have liked E435 but it's strangely not sold in Europe)
    • HP Probook 6475b (can be bought in Sweden but only with A6-4400M so far)
    • Lenovo L430 (this is my backup plan)

Page 1 of 2 12 LastLast

Posting Permissions

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