Quantcast GT650M DDR3 vs GDDR5 difference tested! - Page 11

+ Reply to Thread
Page 11 of 12 FirstFirst ... 789101112 LastLast
Results 101 to 110 of 119
  1. #101
    Notebook Consultant
    Join Date
    Jun 2011
    Location
    new mexico
    Posts
    218
    Rep Power
    5

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    my response.. its a small decrease for diablo3 and other games are either net zero or large gains like batman arkham city. Download it and try it. Bench a few games on the old driver, save the figures in notepad or something, then download and try the new. Share results to compare with others. It helps everyone in the community. Good Luck!
    New and improved m14xR2 - I7, 650 with 2 geebees, some wifi's, SSDs.
    clienterror "Out of curiosity are you one of those guys that buys a BMW but has the crapiest engine and no options with roll up windows just to say you have a BMW?"

  2. #102
    Notebook Enthusiast
    Join Date
    May 2012
    Posts
    11
    Rep Power
    3

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    ok, i have done a bit ocing, only memory, from 2000mhz to 26000mhz
    Unigine benchmark settings:
    Code:
    Render:		direct3d11
    Mode:		800x600 8xAA windowed
    Shaders:	high
    Textures:	high
    Filter:		trilinear
    Anisotropy:	16x
    Occlusion:	enabled
    Refraction:	enabled
    Volumetric:	enabled
    Tessellation:	normal
    GPU: 835MHz(default) Shaders: 1600MHz(default) Memory GDDR5:2000MHz(default)
    Results:

    Code:
    FPS:	36.3
    Scores:	913
    Min FPS:	19.4
    Max FPS:	80.4

    GPU: 835MHz(default) Shaders: 1600MHz(default) Memory GDDR5:2600MHz(oced)


    Code:
    FPS:	37.6
    Scores:	946
    Min FPS:	20.5
    Max FPS:	85.8

    I used msi afterburner to crank up mem to 2600mhz.
    I tried to oc gpu core, but msi afterburner is limiting me to 540mhz, wierd....

    so there you have it, 30% increase of mem freq. results only in 4% performance yield.
    I wish i could oc gpu core to 900mhz, because at these setting temp was below 70C

  3. #103
    Notebook Evangelist
    Join Date
    May 2010
    Location
    UK
    Posts
    612
    Rep Power
    10

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Am I right in thinking that the GDDR5 is only important at high resolutions i.e. 1080p? My new laptop's native res will be 1366x768. Bandwidth isn't as important at this resolution, am I right?
    Samsung 550P5C series:
    Intel Core i5-3210m (@2.5 - 3.1 GHz), 6GB DDR3 RAM (@1600MHz), NVIDIA GT 650M 2GB/Intel HD 4000, Crucial M4 256GB SSD, 1366x768 LED Matte display, Windows 8 Pro 64bit

  4. #104
    Banned
    Join Date
    Mar 2012
    Posts
    865
    Rep Power
    0

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Quote Originally Posted by Cakefish View Post
    Am I right in thinking that the GDDR5 is only important at high resolutions i.e. 1080p? My new laptop's native res will be 1366x768. Bandwidth isn't as important at this resolution, am I right?
    I think gddr5 and 256bit cards are more suited for 1080p but 128bit easily does 720p gaming especially a gt 650m.

    Computer Games on Laptop Graphic Cards - Notebookcheck.net Tech

    Judging from this a gt 650m will play every game you get at 720p at least 30fps and with aa turned down a little most games should play 60fps.

  5. #105
    R3d
    R3d is offline
    Notebook Deity
    Join Date
    May 2006
    Posts
    1,999
    Rep Power
    26

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Right. Assuming all other settings are the same, higher resolution = higher bandwidth usage.

  6. #106
    Notebook Retailer
    Join Date
    Mar 2005
    Location
    Edinburgh, Scotland.
    Posts
    11,760
    Rep Power
    70

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Quote Originally Posted by dado023 View Post
    ok, i have done a bit ocing, only memory, from 2000mhz to 26000mhz
    Unigine benchmark settings:
    Code:
    Render:		direct3d11
    Mode:		800x600 8xAA windowed
    Shaders:	high
    Textures:	high
    Filter:		trilinear
    Anisotropy:	16x
    Occlusion:	enabled
    Refraction:	enabled
    Volumetric:	enabled
    Tessellation:	normal
    GPU: 835MHz(default) Shaders: 1600MHz(default) Memory GDDR5:2000MHz(default)
    Results:

    Code:
    FPS:	36.3
    Scores:	913
    Min FPS:	19.4
    Max FPS:	80.4

    GPU: 835MHz(default) Shaders: 1600MHz(default) Memory GDDR5:2600MHz(oced)


    Code:
    FPS:	37.6
    Scores:	946
    Min FPS:	20.5
    Max FPS:	85.8

    I used msi afterburner to crank up mem to 2600mhz.
    I tried to oc gpu core, but msi afterburner is limiting me to 540mhz, wierd....

    so there you have it, 30% increase of mem freq. results only in 4% performance yield.
    I wish i could oc gpu core to 900mhz, because at these setting temp was below 70C
    Good to know 800x600 is not bandwdth limited by ddr3 *facepalm* i think we figured that out back in the 9800pro/5900ultra days.
    Sales@PowerNotebooks.com | Phone: (US) 877-711-1777 - CALL US FOR NOTEBOOKREVIEW ACTIVE MEMBER DISCOUNTS

    Custom Built Sager, MSI, PowerPro, ASUS and Alienware Laptop computers, and Custom Built Crown Desktop Computers
    ResellerRatings.com Customer Satisfaction Rating =
    10/10 Better Business Bureau A+ Rated

  7. #107
    Notebook Enthusiast
    Join Date
    May 2012
    Posts
    11
    Rep Power
    3

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Quote Originally Posted by Meaker View Post
    Good to know 800x600 is not bandwdth limited by ddr3 *facepalm* i think we figured that out back in the 9800pro/5900ultra days.

    Doing benchmark on FullHD while performing 10fps doesn't make any sense either, so maxed values like 8xAA and Anisotropy 16x impact bandwidth in adequate way for this kind of benchmark software.
    Instead of writing *facepalm* comments, you could contribute with a valid suggestion how to improve testing, so all community would benefit.

  8. #108
    Notebook Guru
    Join Date
    Dec 2010
    Posts
    61
    Rep Power
    7

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    Quote Originally Posted by dado023 View Post
    Doing benchmark on FullHD while performing 10fps doesn't make any sense either, so maxed values like 8xAA and Anisotropy 16x impact bandwidth in adequate way for this kind of benchmark software.
    Instead of writing *facepalm* comments, you could contribute with a valid suggestion how to improve testing, so all community would benefit.
    Unfortunately I'm with the guy above your post... 800x600 isn't a great resolution to bench memory..... that doesn't mean your wrong though how about you prove what you just said to us as I'm very interested...

    Please do another test with AA and Anisotropy off at same Res then two at a higher resolution (720 or 1080p) with AA/anisotropy on/off so we can be the judge of what really affects memory bandwidth.

  9. #109
    Notebook Geek
    Join Date
    Jun 2012
    Posts
    88
    Rep Power
    3

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    I am sorry if that has already been answered but I somewhere read that the GDDR3 version of the GT650M needs more power for the memory than the GDDR5 version. So, is that true and how much warmer will the GDDR3 version get compared to the GDDR5 version?

  10. #110
    R3d
    R3d is offline
    Notebook Deity
    Join Date
    May 2006
    Posts
    1,999
    Rep Power
    26

    Default Re: GT650M DDR3 vs GDDR5 difference tested!

    No. If anything, it needs less power.

 

 
Page 11 of 12 FirstFirst ... 789101112 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
  •  
All times are GMT -4. The time now is 09:35 AM.
Powered by vBulletin® Version 4.2.0
Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.
SEO by vBSEO 3.6.0