Post your upgrade strategies here

When building my pc i was on a budget, so i found components that recieved top reviews around 2 years before but now were around a quarter of the price. I got a athlon 2500 Mobile CPU and overclocked it to save even more.... I will probably upgrade after 3 or so years.

i often see the average pc buyer forking out for a brand new £1500 pc say every three to five years. this means that they are spending an average of £300 to £500 per year and the pc is on average 2.5 years behind the cutting edge.

I think a better strategy for them is to buy 1 year old systems and upgrade every two to three years. This way the pcs will be 1.75 years on average behind the cutting edge and it will only be costing £250 - £375 per year.

This is assuming the price of computing power is halving in price every 12 months.

Obviously this advice would not be suitable for a professional CAD operator who may be paid per hour and would be losing money with a slow system.

Please post up your personal buying strategy along with your reasons...
7 answers Last reply
More about post upgrade strategies here
  1. I generally build high end systems, ~$3000/year. Then I sell them a year later for about half the price, keeping anything still of value for my next build. Although I spend at least $1500/year, I am always using a nice PC.

    Looks like I have a good upgrade path this time though, but I've thought that before, we'll see what happens....
  2. Upgrade Strategies really depend on what the PC is designed for.

    Like you mention about CAD, not only the CPU is one factor, also the GPU/video memory/amount & speed would play another part, as well as what resolution the user uses.

    Upgrading itself is limited by the MB. So to me, that would perhaps be the most important thing to look at.

    So any strategy on upgrading would be in the users preference. Not everyone will benefit on just one particular strategy when cost would be another limiting factor.
  3. Your comment on a CAD operator really gets to the heart of the matter.
    Each of us needs to build a system based on needs and how we intend to use it. Budget is usually an issue, but a hardcore gaming enthusiast will simply have to spend more than an average user to get the performance needed.
    I tend to build my own systems based on quality midrange cost parts that will last for years. This fits my modest requirements for processing speed and 3D rendering. For example:
    Large cheap SOHO tower cases - good cooling, anything fits, indestructible.
    Motherboard with as much on-board utility as possible - saves money, energy, noise and heat, also can use expansion slots for upgrades.
    Quality power supply - I gladly pay extra to keep the power gremlins away year after year.
    Multiple low speed fans - quiet yet effective cooling but does cost extra.
    Latest generation (at the time) processors at the slow end of speed offerings - Obviously saves money but also runs cooler. A 2.4 GHz Northwood or X2 3800+ are examples. Cooling them can be done almost silently.
    Plenty of brand name memory even if it is value ram. I usually do not overclock systems, but have never had a problem with products like Kingston value or Corsair XMS memory. Yes, 2GB costs extra but the swap file never slows me down.
    Video: I will try to buy on-board video if possible, but it is not always a viable option. Second choice is a passively cooled card. Third is the quietest fan cooled card I can get.
    Opticals: 1 dirt cheap DVD-ROM for the boot/source copy drive. 1 high-quality (BenQ or Plextor) burner.
    Floppy: Gotta have one for BIOS upgrades, but that's just me...

    Once built, I will upgrade a system only when the software I want to use just won't work well.
  4. I think you're right. Another thing to bear in mind is that fresher techs are more unstable, so by being 1 year behind the cutting edge, you'll not only get everything cheaper but it will be better too.

    But that's a lot of waste, isn't it? It's not really an 'upgrade strategy' so much as a get-a-new-computer strategy!
  5. Quote:
    I think you're right. Another thing to bear in mind is that fresher techs are more unstable, so by being 1 year behind the cutting edge, you'll not only get everything cheaper but it will be better too.

    But that's a lot of waste, isn't it? It's not really an 'upgrade strategy' so much as a get-a-new-computer strategy!
    Usually the most unstable component is from Microsoft...
    The initial roll-out of new hardware also often has software issues. See the initial reviews of 680i and P965 motherboards as examples. This is short-lived as BIOS upgrades are quickly issued.
    Even using my method of buying a lower-end cpu, I find that I don't need more speed until the next generation (or two) of processors is available.
    This means that at a minimum cpu and mobo go at the same time. Ram and video are the other usual suspects, but may have been already upgraded.
    Today, my wife is still using an Intel D845/P4 2.4 that I upped to 2GB of ram and a 6600 videocard last year. It does everything she needs very well.
    If I ever decide to use Vista, I will swap boxes with her and gut the old Intel. Even this upgrade won't cost more than a grand...

    BTW, did you move?
  6. I upgrade each of my two PCs about every three years. My primary rig is generally more powerful than my HTPC. I try to keep a budget of $1,500 for each PC excluding the cost of a monitor upgrade, but that is not a hard rule. I also try to keep as much hardware from the previous build as possible. That's usually the hard drive, sound card, and DVD drive.

    My old primary rig was built around an Athlon XP-M (which I gave to my cousin). My aging HTPC is also built around the same CPU. Since Blu-Ray and HD DVD movies requires a pretty powerful CPU (E6600 is the minimum recommendation), the HTPC will be at least as powerful as my primary rig with the exception of the GPU since I don't play games on my TV.

    I may actually take the opportunity to move my C2D E6600 from my primary rig and place it in my upcoming HTPC and drop in a more powerful CPU into my primary rig. Otherwise, I'll just buy another E6600.
  7. i buy the absolute best i can afford and run it at stock. some time later (around 12 months typically) when i get the "its too slow" feelng, i overclock it. then i start the process all over when the "slow" feeling sets in again (another 12 months typically).

    this strategy typically gets me 2-3 years out of a computer for $1000-$1500, but i get the feeling of a new computer every year or so :)
Ask a new question

Read More

Homebuilt Systems