Upgrading from a Radeon x1600 pro g.c.

  1. #1
    jamieren is offline Newbie

    Upgrading from a Radeon x1600 pro g.c.

    I've had this graphics card for over a year now and can admit to myself I've never been overly impressed with it. Starting with the $160cdn price point (I think I got took) and the fact that it gave moderate framerates (20-30 at best) at medium graphical settings ..my favorite games being Stronghold: Legends and X3: Reunion. I'm looking at upgrading, and it seems the Radeon HD 4830 would be the best choice for me, according to this website:

    YouGamers - Hardware - 3DMark06 User Hardware Statistics - Price & Performance

    I'm not excited about dropping another $100 or more on a video card, but what else to do? It seems to spend less would only result in letdown. I currently am running 1 gb RAM (but am planning to double that for a measly $20 U.S.). My main concern is this talk of PCI-E 2.0. The 4830 card looks identical to my x1600 as far as the pins are concerned, but will it fit on my motherboard? My assumption is the 2.0 part refers to the ability to run 2 cards concurrently, and thus requiring 2 PCI-E ports, right? I do not believe my computer has two ports (nor will I need them with only one card) ..so have I answered my own question or what?

    Additional info: 2.2 gig 3500 athlon processor (dual core?), computer model SR1750NX (Compaq Presario).
    Also, if anyone can direct me to a different decent performing card (particularly if the 4830 will not work with my computer!), I'd be happy to hear your advice!
    *EDIT* I see this card is on sale @ tiger direct for $80 U.S.
    Last edited by jamieren; 28-03-2009 at 05:58 AM. Reason: clarification


  2. #2
    jephree is offline ¨*·.¸ «.·°·..·°·.» ¸.·*¨
    Running two cards with ATI is known as CrossFire.

    Running two cards with NVIDIA is known as SLi.

    This has nothing to do with PCI_E 1 | PCI_E 2

    The PCI_E spec depends on your motherboard. Most new high end cards are PCI_E 2 but are also backwards compatible to PCI_E 1 although their full features will only be available under a motherboard rated at PCI_E 2. So this distinction is a motherboard issue. Not a graphics card issue. The slot is the same in both cases. The performance is not. For full performance of a PCI_E 2 card you need a PCI_E 2 capable motherboard.

    As to my own preference I only work with NVIDIA. From a long past history I don't even consider ATI / AMD. But that is just my acknowledged prejudice here.

    2GB of RAM will make a nice difference whatever card you use.


    ............
    Last edited by jephree; 28-03-2009 at 06:39 AM. Reason: typo

  3. #3
    Digerati is offline Senior Quiquagenarian
    I suggest checking out Best Graphics Cards For The Money: March '09 : February Review/March Updates - Review Tom's Hardware - then make sure your power supply can support your new, more power hungry card.

    I agree on another point with jephree - bumping up your RAM to 2Gb will make a nice difference, regardless the card. I recommend adding RAM first - it will provide an overall performance boost that may make you happy with your current graphics card.

  4. #4
    jamieren is offline Newbie
    Quote Originally Posted by jephree View Post
    For full performance of a PCI_E 2 card you need a PCI_E 2 capable motherboard.

    As to my own preference I only work with NVIDIA. From a long past history I don't even consider ATI / AMD. But that is just my acknowledged prejudice here.

    2GB of RAM will make a nice difference whatever card you use.
    ............
    Thanks for the clarification. I'm assuming my motherboard is PCI-E '1', I picked this thing up in 2006. I will most definitely install the RAM before making any decisions about video cards with this info. I've got 'sources' (computer savvy salesmen, really) that told me bang for buck, ATI almost always outperforms Nvidia -although you would feel otherwise I suspect; and I may be inclined to believe you. I also thought that perhaps because Radeon is related to AMD (and my computer has an athlon processor) the two might mesh better, no?

    One more thing. I'm using 'mad onion's' 3Dmark2001 benchmark (I know it's outdated, but still gives a reference point for performance without pushing the computer to the breaking point). Well, I had this installed on my computer before (I switched to Vectorlinux for about a year) and the computer benched around 15000 points. Then (a 'friend') screwed up windows ..hence the change to linux. Now I did a system restore and downloaded the latest drivers/ windows programs to update & help make the unit run more efficiently. I benched last night and the result was 11000. Here's the kicker though: I watched the display and saw the machine was *consistently* running anywhere from 15 to 80 fps faster than before at different points!! How is this possible if the benchmark actually gave it a lower rating?? My first impression is to take the framerate & run, bench scores be damned! Seriously though, anyone have an explanation?

  5. #5
    Digerati is offline Senior Quiquagenarian
    I've got 'sources' (computer savvy salesmen, really) that told me bang for buck, ATI almost always outperforms Nvidia
    Well, the defining word in your statement above about those guys is they are "salesmen". It is an invalid, blanket statement. From a consumer's standpoint, the bottom line is based on (1) Primary function/purpose of the computer and (2) the budget - then you go for the card that achieves the best balance. At some price points, it is ATI/AMD. In others, it is nVidia. Then next year, it might be reversed.

    As for benchmarks - I really don't pay too much attention to them - unless it is benchmarking a specific task that is important to me, and how I use my computer. And even then, that is only while researching professional reviews (unbiased, in a lab, using scientifically obtained, empirical data) to make my purchasing decisions. Once I get the product installed, up and running, and doing the tasks it was purchased for, I don't need a benchmark to tell me if the computer is running up to snuff or not.

    My point is, benchmarks results, and results you "perceive" in the real-world can, and often are, entirely different. And because of the Placebo Effect, when benchmarks go up or down, our expectations affect our perceptions - not always a good thing.

    If it looks good, leave it alone.

+ Reply to Thread