Sponsor Message:
Non Aviation Forum
My Starred Topics | Profile | New Topic | Forum Index | Help | Search 
PC Video Card: How Much To Notice A Difference?  
User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
Posted (8 years 7 months 2 weeks 5 days 3 hours ago) and read 1858 times:

What I have now in a laptop:

  • Intel Extreme integrated graphics, 64megs


  • What I plan to get for a future desktop replacement:
  • SAPPHIRE Radeon X1600PRO, 512MB, PCI x16


    • What can I expect from having had Integrated 64 megs up to this point?
    • Will it make a difference as to what I'm getting, i.e. do I need 512Megs?
    • What are pixel pipe lines?
    • What is the meaning in the core gpu and memory clockspeeds?



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    23 replies: All unread, jump to last
     
    User currently offlineLHMARK From United States of America, joined Jan 2000, 7255 posts, RR: 46
    Reply 1, posted (8 years 7 months 2 weeks 5 days 3 hours ago) and read 1854 times:

    What are you doing with the laptop? Are you playing games and/or editing images and video?


    "Sympathy is something that shouldn't be bestowed on the Yankees. Apparently it angers them." - Bob Feller
    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 2, posted (8 years 7 months 2 weeks 5 days 3 hours ago) and read 1849 times:

    I've tried to play games with it (resulting in a loud & hot processor fan) and watch dvd's. I dont edit anyting at this time. I would like a tv tuner maybe...


    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineAirWillie6475 From United States of America, joined Jan 2005, 2448 posts, RR: 1
    Reply 3, posted (8 years 7 months 2 weeks 5 days 3 hours ago) and read 1845 times:

    DONT WASTE YOUR MONEY on a nice $300 graphics card if you don't have the updated system. Make sure you have the latest Pentium chip, at least 512mb of ram, and a PCI express slot in the motherboard which the latest graphcis card technology. Without these, I don't care if you have the best card in the world, it will not perform well. Clockspeed is like the Pentium chip of the PC except it's only for your Graphics card. As with the Pentium, the higher the number on the clock speed the better. It's the brain power of the graphics card.

    Just to give you an idea of my case. I have an 2.8 Celeron chip, with 752mb of RAM, and a ATI card with shared 256mb. The highest graphics setting I can set on FS2004 and fly comfortably is medium high. I could probably set it a lot higher with my graphics card but the Celeron chip won't allow it. In conclusion the Processor is the most important thing when it comes to your graphics performance.


    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 4, posted (8 years 7 months 2 weeks 4 days 20 hours ago) and read 1834 times:

    "Integrated graphics", also identified as "shared memory" is only used on low-end Windows laptops; It incurs a severe performance penalty. When you're switching from such a system to one which has a proper graphics processor with dedicated video memory (not shared with the CPU), there will be a very noticeable difference, especially with anything 3D, but probably also with normal window handling.

    I'm not up-to-date on graphics processors, but as far as I'm aware the X1600 is a pretty decent one; 512MB dedicated video RAM is quite good, 16x PCIe is also top-of-the-line.

    I'd rather check for the noise level the card produces - a noisy graphics card can be very annoying (especially with those tiny fans with their high-pitched whine) and may not be worth the increased GPU clock frequency; Unless you're planning major high-end gaming, a fan-less (and somewhat slower clocked) graphics card may be a worthwile alternative. In any case it is a good idea to listen to the system you're about to buy before making the decision.


    User currently offlineEilennaei From , joined Dec 1969, posts, RR:
    Reply 5, posted (8 years 7 months 2 weeks 4 days 18 hours ago) and read 1815 times:

    If your system is recent enough to have a PCI Express slot, your CPU is up to the task. The only improvement you'll see is in gaming and other 3D. Normal office applications will not improve noticeably. The 512 MB onboard only refers to a space to hold 3D "raw information" on the graphics card. When you install the new card, your system will be smart enough to revert all of the RAM to the applications from the dual video/application use.
    The speeds refer to the internal operation speed of your video card. The same basic chip can be run with different speeds with different memory chips. Overclockers love to tweak these values very close to the limit where the 3D card will intermittently crash (utilities are often submitted with the card).
    The fan on the card seems to be quite large in diameter, and hopefully quiet. Fanless cooling system kits are available that fit most cards .

    [Edited 2006-02-01 14:51:16]

    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 6, posted (8 years 7 months 2 weeks 4 days 18 hours ago) and read 1806 times:

    Quoting Eilennaei (Reply 5):
    The only improvement you'll see is in gaming and other 3D. Normal office applications will not improve noticeably.

    Relative to shared memory graphics, the improvement should still be quite noticeable.

    Quoting Eilennaei (Reply 5):
    Normal office applications will not improve noticeably. The 512 MB onboard only refers to a space to hold 3D "raw information" on the graphics card.

    No, the video memory holds the frame buffer (the pixels you currently see on screen) and - depending on the operating system - also window backing buffers, text character caches, video buffers and many other graphics-related items. The difference may not be as pronounced under older systems (such as Windows XP or older), but MacOS X and also presumably Windows Vista make extensive use of the video memory available in order to offload as much graphics work as possible on the graphics processor.


    User currently offlineEilennaei From , joined Dec 1969, posts, RR:
    Reply 7, posted (8 years 7 months 2 weeks 4 days 13 hours ago) and read 1787 times:

    No, the standard office application performance will not be boosted significantly by the addition of a separate 3D graphics card. The basic office PCs are commodity items, whose performance and price is matched to the standard office applications. If a separate 3D processor helped greatly vs. the price, it would have been installed already. Moreover, the "shared mem" systems do include some 3D primitives.

    When I said that a "512 M on the card" only refers to the 3D performance, I meant just that. A mere 8 MB is enough to hold the normal 2D frame buffer, and increasing card memory to 64M, 128M and so on will not improve 2D.

    (Klaus, whenever I write about PCs, I tend to omit things that I think are of no importance to the standard user. That does not (always) mean that I do not know how things stand.)

    For Klaus' reference on one shared mem system architecture:
    http://www.intel.com/design/chipsets/applnots/30262403.pdf


    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 8, posted (8 years 7 months 2 weeks 4 days 11 hours ago) and read 1778 times:

    Quoting Eilennaei (Reply 7):
    No, the standard office application performance will not be boosted significantly by the addition of a separate 3D graphics card.

    As far as I understand, Lehpron is planning to replace his old, underpowered, shared-memory laptop with a completely separate desktop system which also contains the PCIe card he described above.

    Quoting Eilennaei (Reply 7):
    If a separate 3D processor helped greatly vs. the price, it would have been installed already.

    No. Bare-bones cost-cutting and the customers' willingness to sacrifice quality for a few bucks "saved" on acquisition price have led to the creation of the crippled "shared memory" architecture in the first place, so that is what it's being used for.

    Quoting Eilennaei (Reply 7):
    Moreover, the "shared mem" systems do include some 3D primitives.

    3D operations are just one of many uses modern GPUs have for a system. Whenever you're scrolling, moving or resizing a window or when you're displaying text or video, the GPU is being used - and correspondingly slowed down in case of shared memory.

    It gets a lot tougher when advanced compositing mechanisms are being used for window display as under MacOS X (and soon under Windows Vista). No 3D-operations required anywhere, and still GPU performance is very relevant to overall system performance.

    Quoting Eilennaei (Reply 7):
    When I said that a "512 M on the card" only refers to the 3D performance, I meant just that. A mere 8 MB is enough to hold the normal 2D frame buffer, and increasing card memory to 64M, 128M and so on will not improve 2D.

    Not true at all. The impact may be somewhat smaller than with a modern operating system, but it is still there. Moving lots of pixels fast is a prime function of the GPU, so a decent graphics card is increasingly important even for "normal" non-3D use.


    User currently offlineEilennaei From , joined Dec 1969, posts, RR:
    Reply 9, posted (8 years 7 months 2 weeks 4 days 10 hours ago) and read 1772 times:

    Quoting Klaus (Reply 8):
    Quoting Eilennaei (Reply 7):
    If a separate 3D processor helped greatly vs. the price, it would have been installed already.

    No. Bare-bones cost-cutting and the customers' willingness to sacrifice quality for a few bucks "saved" on acquisition price have led to the creation of the crippled "shared memory" architecture in the first place, so that is what it's being used for.

    Klaus, you need to understand not everyone needs the top-of-the market system. Managers acquiring 1000s of computers will look into every penny.
    You may think all people in the acquisition process stupid, suits me, no problem.


    Quoting Klaus (Reply 8):
    3D operations are just one of many uses modern GPUs have for a system. Whenever you're scrolling, moving or resizing a window or when you're displaying text or video, the GPU is being used - and correspondingly slowed down in case of shared memory.

    Klaus, seems you're prone to fixations, unfortunately. If you had read the link I posted, you'd be shown that the data bandwith of the "crippled" GPU system can in fact be quite impressive.


    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 10, posted (8 years 7 months 2 weeks 4 days 9 hours ago) and read 1762 times:

    Quoting Eilennaei (Reply 9):
    Klaus, you need to understand not everyone needs the top-of-the market system. Managers acquiring 1000s of computers will look into every penny.
    You may think all people in the acquisition process stupid, suits me, no problem.

    Corporate commodity acquisition has certainly played its role in lowering overall quality standards, but that is only marginally relevant here. Lehpron iis buying for himself, so quality is an issue he'll have to live with one way or another. Hence "shared memory" should be avoided if possible.

    Quoting Eilennaei (Reply 9):
    Klaus, seems you're prone to fixations, unfortunately. If you had read the link I posted, you'd be shown that the data bandwith of the "crippled" GPU system can in fact be quite impressive.

    You're remembering graphics card issues as they once had been (GPUs only being used for 3D in practice), but things have already changed, and the change is progressing.

    The chipset described by the document you linked to has a total peak bandwith of 8.5GB/s. That means a) sustained typical bandwidth is much lower and b) CPU and GPU have to share that bandwidth.

    In effect both are falling over each other's feet at every turn and are reducing each other's performance. In a "proper" system the GPU has much more than that for its dedicated video memory alone!


    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 11, posted (8 years 7 months 2 weeks 4 days 7 hours ago) and read 1754 times:

    I thank everyone for your replies, especially Klaus and Eilennaei for entertaining my brain. I need to be informed.

    Quoting Klaus (Reply 8):
    As far as I understand, Lehpron is planning to replace his old, underpowered, shared-memory laptop with a completely separate desktop system which also contains the PCIe card he described above.

    Hit nail on head. I already bought a the desktop, it's got a gig of memory DDR2 and dual core 2.8GHz CPU. Will not arrive til Valentine's day...

    Quoting Klaus (Reply 10):
    Lehpron is buying for himself, so quality is an issue he'll have to live with one way or another. Hence "shared memory" should be avoided if possible.

    Lesson learned believe me. At the time, I just wanted a laptop for the laptop. I ended up using software with twice the reqs I had, for the most part. Then comes the GDDR2/GDDR3, pixel pipelines and gpu/memory clockspeeds, and I'm like CONFUSED.

    Quoting Klaus (Reply 8):
    No. Bare-bones cost-cutting and the customers' willingness to sacrifice quality for a few bucks "saved" on acquisition price have led to the creation of the crippled "shared memory" architecture in the first place, so that is what it's being used for.

    From my experience, I think the cost-cutting on my part was ignorance due to what was good for me and what I could afford. Then and now, computer form factor was an issue, so I could stick the thing in my siotecase when I go to school. Looking back for the price I paid, I could have gotten a Dell 4700C which was a bit better. It had up to 128MB dedicated video ram. But I do not know how to standardize this.

    It must be how AMD chips are not about their CPU clockspeed, that graphics cards are not entirely about the amount of onboard video memory?

    Quoting Klaus (Reply 10):
    Corporate commodity acquisition has certainly played its role in lowering overall quality standards, but that is only marginally relevant here.

    While probably irrelevant, my new computer was from the small business section rather than the home & office section. Size was an issue, I wanted the damn thing to be as small as possible yet powerful and expandable. I'm quite surprised I got the deal I got!



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 12, posted (8 years 7 months 1 week 3 days 14 hours ago) and read 1729 times:

    I just got my new computer yesterday and while I was disapointed that it only takes low-profile (LP) add-on cards, the speed of this thing facinates me, it's quite blistering despite integrated GMA 950.

    Nevertheless, I still need a video card, but now my choices are limited, severely. Newegg.com has the widest selection of computer hardware online and it is where I found these two, what i call "best bets" as of yet:

    Radeon X700, 256MB DDR, PCI x16
    Geforce 7300GS, 256MB (with 512 turbocache support) GDDR2, PCI x16

    Both are low profile but neither have LP brackets, which ticks me off, as it will not fit. I hope these brackets can be removed as I can't even find the brackets online. That Geforce card uses a combination of shared and onboard memory to operate. Nvidia claims it is 4x faster than shared off the motherboard but doesn't compare to a dedicated vid-memory card of same size. The core clock of the Nvidia is 550MHz while the 400MHz for the ATI card in addition tot he fact that the latter is more expensive. Why? The ATI card has 8 pixel pipelines and the Nvidia card has 4. I still do not know what they are/mean but I equate them to fliud pipes. They are both the most powerful (for my uses) LP cards I could find online. Brick and mortar stores do not seem to be selling them.

    I have tried contacting both companies including those partners that make the cards, I'm in a holding pattern. I can't find too many resources on those two cards, so I am asking you folks for anything on your minds.

    Had I been able to fit a full height card, I would have gone with neither ATI or Nvidia. I would have gone with S3 Graphics; I was sold on their S27 and I pray for an LP version. What I see as their moto is efficiency in terms of performance per watt.



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 13, posted (8 years 7 months 1 week 3 days 10 hours ago) and read 1717 times:

    It seems you're in a real pickle with the low profile slots... Not a good thing. Sad

    From a superficial look at the specs, it seems the Geforce may be weaker one of the two, since it has the narrower (and thus slower) memory interface at 64 bits vs. 128 bits for the Radeon. Double the pixel pipelines would normally make a positive difference in 3D-performance as well. A relatively minor difference in clock speed is probably less relevant.

    In the end the actual performance can depend on many factors, so a real comparison test would be more helpful here.

    To my knowledge ATI and Nvidia are the only manufacturers who are still seriously in the game, with all other manufacturers having basically dropped out of the main performance competition and trying to survive in marginal niches of their own.

    Okay, so let's see if anybody else can contribute any more concrete information...


    User currently offlineEilennaei From , joined Dec 1969, posts, RR:
    Reply 14, posted (8 years 7 months 1 week 3 days 9 hours ago) and read 1704 times:

    Quoting Klaus (Reply 10):
    In effect both are falling over each other's feet at every turn and are reducing each other's performance. In a "proper" system the GPU has much more than that for its dedicated video memory alone!

    Did I mention fixation? As I said earlier, the net result is good enough vs. the price. Here's an actual comparison between a shared mem vs "true" system:

    230 / 211 SysMark2002, a penalty of 9%
    166 / 150 SysMark Productivity Rating, a penalty of 10%

    For that small performance penalty, you can drop the separate GPU and its video memory and other circuitry from the computer. It's a purely economic matter for a mass-produced computer. I'm sorry if you fail to grasp the idea that not everyone needs and/or can afford a top-of-the line computer.

    http://reviews-zdnet.com.com/NetVista_S42/4514-3118_16-20584107.html


    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 15, posted (8 years 7 months 1 week 3 days 7 hours ago) and read 1698 times:

    Quoting Eilennaei (Reply 14):
    Did I mention fixation?

    Repeatedly!

    Actually, it does look a bit like a fixation to me...!  mischievous 

    Lehpron is not a corporate buyer trying to pinch the last penny out of a purely office-destined computer with a pre-limited lifespan.

    How about sharing some of your wisdom regarding his actual questions?


    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 16, posted (8 years 7 months 1 week 2 days 8 hours ago) and read 1670 times:

    Quoting Klaus (Reply 13):
    From a superficial look at the specs, it seems the Geforce may be weaker one of the two,

    So, again from superficial, would the potential 512 megs total (onboard + shared) be irrelevant? In addition, the Geforce's memory type is GDDR2 and the Radeon's memory type is DDR, difference?

    Personally I am quite surprised at the performance of what looks like 128meg shared memory, its only double what i had on my laptop. I ran the game of HALO on the highest settings on the easiest difficulty and died for the first time. But Age of Empires 3 still lags on lowest settings. DVD's and music runs like charms. I do not know what is doing it though, either the CPU/graphics abilities...?

    Still, I thought Intel945G had 224megs shared on the board, not 128, I'm going to call up the company to ask what's up -- could it be a bios thing?

    Speaking of performance, benchmarks like 3Dmark, can they give an approximate of what is needed to run a program? I'll see what I had now to compare...FWIW, my laptop scored in 3Dmark2003 as 1000, after which my drive crashed. Big grin This was two years ago.



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 17, posted (8 years 7 months 1 week 2 days 8 hours ago) and read 1664 times:

    It's a whole science unto itself to evaluate graphics card performance, especially regarding the effective performance in games. Some games can be CPU-limited, so a faster CPU will get you an improvement, while even the fastest graphics card wouldn't. Others are limited by number of pixels moved, by number of objects on the screen at the same time, the number of textures used, etc...

    Graphics card memory only becomes relevant when the existing amount is being filled / exceeded. Up to that point more memory won't make a difference. I can't say if the ability to access another (slower) 256 MB of RAM may actually make a difference in games you'd use.

    But a narrower data bus and a lower number of shader pipelines look like a real handicap. Still, I'd try to find some actual comparison tests.


    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 18, posted (8 years 7 months 1 week 21 hours ago) and read 1641 times:

    Well I found a GeForce equivalent to the Radeon X700...in the end it may come down to price.

    Quoting Klaus (Reply 17):
    I can't say if the ability to access another (slower) 256 MB of RAM may actually make a difference in games you'd use.

    Granted, but not just for games. I do recall quite quickly the difference more shared memory made when comparing a program my laptop and the same on the new computer, I cannot imagine the performance jump (if any) an independent memory will do... I have always figured using system memory was a bad thing as it kept from other uses. It was always one or the other, multi-tasking was like a cliche or a hype that didn't apply to me.

    While I have a dual-core CPU, some things are still slow, I suspect that I am still depending on shared memory. We'll see, I'm in no hurry to make a purchase, just necessary for projects I have in mind.

    Is it true what I hear about having a second empty HDD with a vast paging file that can act like a large memory source?



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineKlaus From Germany, joined Jul 2001, 21461 posts, RR: 53
    Reply 19, posted (8 years 7 months 1 week 16 hours ago) and read 1628 times:

    Having a separate harddisk just for paging can indeed improve performance - but only if you don't have enough RAM. The hierarchy is:

    a) Enough RAM: By very, very far the most efficient alternative is to avoid swapping all together during normal use.

    b) Separate swapping disk: Orders of magnitude slower than the no-swapping case, but still somewhat faster than using a single (usually also fragmented) disk for both storage and swapping.

    c) Worst case: Too little RAM and only one disk for both storage and swapping. Can be twice as slow or worse than a separate swapping disk, but the difference still pales in comparison to a proper amount of RAM avoiding swapping in the first place.


    To the graphics card: As far as I've seen, the Geforce would start using shared memory if its own dedicated video RAM becomes full, so performance degradation may be less than being limited to just the dedicated RAM and having to "manually" re-load it. Unfortunately, the dedicated memory has a relatively weak interface (only half as wide as the Radeon), so that looks like a bottleneck even in the best of cases.

    But again, you should try to find some actual tests, preferrably for the kinds of uses you're intending.


    User currently offlineEilennaei From , joined Dec 1969, posts, RR:
    Reply 20, posted (8 years 7 months 1 week 6 hours ago) and read 1617 times:

    Quoting Lehpron (Reply 16):
    Personally I am quite surprised at the performance of what looks like 128meg shared memory

    You likely have in your machine the new generation DDR2 main memory whose data transfer rate is very high. My mate Klaus told us in #10 basically that it's not that high actually, and that such systems are crippled.

    Quoting Lehpron (Reply 16):
    Still, I thought Intel945G had 224megs shared on the board, not 128, I'm going to call up the company to ask what's up -- could it be a bios thing?

    Memory is being dynamically allocated by the chipset:
    http://download.intel.com/design/chipsets/applnots/30750801.pdf
    http://www.intel.com/products/chipsets/945g/prodbrief.pdf

    Will return to the other outstanding issues later today..


    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 21, posted (8 years 7 months 1 week 5 hours ago) and read 1615 times:

    Quoting Eilennaei (Reply 20):
    Memory is being dynamically allocated by the chipset:
    http://download.intel.com/design/chipsets/applnots/30750801.pdf
    http://www.intel.com/products/chipse...f.pdf

    Problem fixed, it was in the BIOS.

    Quoting Eilennaei (Reply 20):
    You likely have in your machine the new generation DDR2 main memory whose data transfer rate is very high.

    I have DDR 533MHz, while I have the option in the future to go for 667MHz, I can instead go upto 4gig's if I wanted too. I like having this room to expand, unlike a laptop.

    Quoting Klaus (Reply 17):
    But a narrower data bus and a lower number of shader pipelines look like a real handicap. Still, I'd try to find some actual comparison tests.



    Quoting Klaus (Reply 19):
    But again, you should try to find some actual tests, preferrably for the kinds of uses you're intending

    I know, I know, dude. There are not too many comparisons of low-profile, half-height cards. So far I am dependent on companies claiming their card competes with blah blah. For example, the Geforce 6600 and Radeon X700 low profile cards have the same DDR2 mem type and 8 pipelines and have around 500 MHz mem clock. Both have dedicated memory, none shared.

    But, the Geforce 7300GS is quite new and appears to be going after the X1300, low profile version. Nvidia has yet to update their site, they claimed the 6200 low profile was to compete with X600SE. the 6200 and 7300 appear to have both shared and dedicated memory, but the 7300 is the only one I've seen has the capability to support 512megs. But as you've said, it may suffer performance degradation. Although, the 7300GS's video memory clock and my system memory clocks are close to the same, would i notice the performance degradation?

    The Nvidia website was the only place online that would straight out tell me that the GeForce 6600 cards were best with the game of AoE 3. The ATI website adn supporting companies have yet to mention what cards work best with whatever. Only because the 6600 low profile is slightly cheaper than the X700 equivalent, I'm leaning in that direction. But, like I said, I'm in no hurry.

    At this time, I'm waiting for what comes down the pipeline, no pun intended. The X7/X8 and GeForce 6-series were high-end cards last year. Now we have the 7-series and X18/X19's. While I won't wait forever (as the whole point to getting a new computer was the video card), I will not wait til we see an 8-series or a Radeon X2200...  Wink



    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    User currently offlineRayChuang From United States of America, joined Jun 2000, 8016 posts, RR: 5
    Reply 22, posted (8 years 7 months 1 week 3 hours ago) and read 1608 times:

    Unless you run a game that really takes full advantage of all that DirectX 9.0c offers, most higher-end graphics cards are ridiculous overkill.

    A graphics card that uses the nVidia GeForce FX 5200 chipset is more than enough for most user needs.  Smile


    User currently offlineLehpron From United States of America, joined Jul 2001, 7028 posts, RR: 21
    Reply 23, posted (8 years 7 months 1 week ago) and read 1603 times:

    Quoting Lehpron (Reply 21):
    I have DDR 533MHz

    I meant I have DDR2 533MHz...   

    FWIW, I did a 3Dmark2001 benchmark test just to give myself and others an idea of what and what this all means (I do not have video card in yet, its all shared at this point) (the dowload of demo software for the 2006 version took too long)

    3DMark2001 Score.................5866 (this is 5 times that of my laptop!)

    Game 1 Car Chase
  • Low Detail 78.8 FPS

  • High Detail 31.1 FPS

  • Game 2 Dragothic
  • Low Detail 82.4 FPS
  • High Detail 42.0 FPS
    Game 3 Lobby
  • Low Detail 112.5 FPS

  • High Detail 38.4 FPS

  • Game 4 Nature
  • 45.0 FPS


  • Fill Rate
  • (Single-Texturing) 1503.7 MTexels/s
  • (Multi-Texturing) 1505.7 MTexels/s

    High Polygon Count
  • (1 light) 9.0 MTriangels/s

  • High Polygon Count
  • (8 lights) 5.3 MTriangels/s


  • Environment Bump Mapping 159.6 FPS
    DOT3 Bump Mapping 115.1 FPS
    Vertex Shader 50.7 FPS
    Pixel Shader 157.8 FPS
    Point Sprite 6.2 MSprites/s

    [Edited 2006-02-13 09:40:18]


    The meaning of life is curiosity; we were put on this planet to explore opportunities.
    Top Of Page
    Forum Index

    This topic is archived and can not be replied to any more.

    Printer friendly format

    Similar topics:More similar topics...
    How Much To Ship A Car Transcontinental? posted Tue Apr 5 2005 08:12:08 by Flybyguy
    How Much To Tip In The USA? posted Thu Nov 18 2004 03:20:30 by AMS
    How Much Work To Get A Pic Like This? posted Sun Jun 18 2006 04:28:25 by Timz
    How Much Do You Want Everton To Win Tonight? posted Wed Feb 8 2006 20:31:55 by Cosec59
    How Can I Get A Girl To Notice Me? posted Wed Jul 13 2005 23:12:09 by Alberchico
    How Much Is Air Conditioning Worth To You? posted Sun May 22 2005 03:44:11 by TheCoz
    Cab From FLL To Miami-How Much? posted Wed Oct 20 2004 04:44:32 by GD727
    How Much Is It Worth To You? posted Sat Aug 14 2004 06:08:33 by ArmitageShanks
    PHL To DC, How Much Toll? posted Sat Aug 14 2004 02:43:10 by Ushermittwoch
    How Much Is It To See A Movie Where You Live? posted Sun Sep 14 2003 18:52:52 by Jcs17