Author: aztpg8

Posted on

Converting a Dell Optiplex…. Interesting.

How did we get here?

I knew there may be more to this than meets the eye when I started, but it’s down the rabbit hole I go. First, I got the Xeon CPU to test. Perfect, so far. Then, after realizing it needed more ‘space’, I ordered an Optiplex 9020 motherboard because it was cheap. The board works beautifully, but there are some challenges.

The first is the power supply connections. I do have an adapter, but the PSU in question doesn’t seem too happy with it. Next is the front panel and power switch connections. Oh brother, more adapters. Lovely. I can do it without adapters, but there is a lot of jerry rigging involved. I went safe, and for only a few bucks here and there, we are still well below the price of a mainstream motherboard.

At what cost

The biggest challenge is staying below what a mainstream board would have cost, but I don’t want to set the house on fire trying to do it. It wasn’t on my list of goals for the week. It’s not on it for next week either, if you were wondering. I do have some parts, but the trick here will be to come in as close to $250 as possible and still have a solid gaming PC. Buying a mainstream motherboard would have killed that, but 20 bucks is solid.

Counting what I have on the shelf, we’re talking around 250 so far. If we ignore that, it’s much cheaper, but defeats the point. Using all of the parts versus buying an Optiplex in a bigger case might eventually still be a good comparison. The results of this experiment will tell me if I should try it. Maybe a standard sized PC and a can of spray paint would be just as effective. I may as well start looking for one to order.

So, what’s next for the Optiplex?

The next part is waiting for the adapters to come tomorrow, and making sure the PSU will work with what I already have. That in itself will be one video. After that, we start testing and see what it best compares to. I know I will compare numbers to the i5 4670 in the slimline, but I also want to compare numbers to a modern four core CPU to see what’s missing after six generations.

The ‘finished’ product will end up on the channel, here, but of course this will be where I discuss what went right and what went wrong, so if you’re reading this, please check back. Until then, I wait. I have begun to assemble the parts that are here, and I probably could do some testing, but then I wouldn’t have needed the parts, and I wouldn’t get to cover the adapters in the video. Six of one, half dozen of the other. It is bound to all work out some way.

Stay tuned, it should be fun.

Return to the Blog page

Posted on

Was it actually an upgrade? Xeon in an Optiplex.

How did we get here?

So, the most recent upgrade to the Dell Optiplex was the addition of an Intel Xeon E3-1270v3. A four core, eight thread CPU built on the Haswell platform similar to the i5 4670 that came with the rig. There had already been a few upgrades, each bringing this modest office PC closer to being an actual budget Gaming rig. The video is found here.

The idea was simple enough, a more powerful CPU without spending a lot of money had the chance to turn it into a bit of a powerhouse. It didn’t work out that way. By all accounts, doubling the amount of cores on the CPU should have made many jobs much easier. Instead we reached a different issue, the Graphics Card.

I tested both CPUs on games and benchmarks at 1080p. It’s the most common resolution in games and should have relieved the GPU of stress. It didn’t. It didn’t actually create any advantage at all. Most tests showed similar results for each processor, leaving no clear winner. But why?

What went wrong?

The biggest limitation is the size of the case. Only certain GPUs will fit in the small form factor, and even the best available isn’t adequate. I say the best available, but I should clarify, the best available at that budget. There are better GPUs, but at a much higher cost, and the difference may not be worth it. The fact is, you can build a cheaper, better performing, new PC for what it would cost.

Heat seemed to be another issue. The set up didn’t overheat, but it did get warm and that wasn’t a help. Incidently, any PC built from new parts will most certainly have better airflow. Temps on the CPU averaged about ten degrees more and the GPU was the same. Heat in small areas is always a problem and this set up is no exception. More, slightly faster threads, work harder and the heat has nowhere to go.

So, is the Xeon a failure?

Failure? Absolutely not. The small form factor has reached a point of dimishing returns, and although the Xeon didn’t improve things, it’s not a loss. The i5 4670 works great in the system, and I’ve ordered a different motherboard for the better processor. I can pair it with a much better GPU in a case with more airflow and give it a chance to “let it’s hair down”. If there is no improvement at that point, maybe we can call it a learned experience, but most cetainly not a failure.

The replacement motherboard is on it’s way. I already have a case, (which was also a bit of a challenge), and memory, etc., so it won’t be hard to set up, Then, we’ll just see how it the CPU does in a better situation. It will be interesting to see the improvement, assuming there is one. Obviously, there will be a video, and a blog, so stay tuned. It promises to be interesting.

Back to the blog page

Posted on

Let’s try a different tack. It’s time to make an important change.

Originally, this website would be to share pictures of builds, recent products and to redirect everyone to my blog on Buy Me A Coffee. With the loss of the Amazon Affiliate partnership, that changed things. It wouldn’t be very effective to only show a few pics and just be a hub where you could find my other socials, so I’m going to make a bit of an adjustment. Actually, I’m going to make a major adjustment. It most likely won’t be the last, but its a start. I guess either different tack, or tact would apply here.

This will now be the home of my tech related blog. The non-tech stuff will stay on BMAC. The goal will be to post weekly and provide some additional thoughs and insight. It will be a chance to give a more in depth analysis that I can share in a ten to fifteen minute video. It also gives me a chance to go more in depth on several subjects. This blog won’t have the chance of being monetized, but that’s not a concern yet.

It will also be helpful to explain my methods and why certain things don’t get covered. If I’m using higher settings or more tests, or the opposite, I can explain why. It may also be a way to explore more subjects for videos. At the very least, I’ll use it to fill out what gets missed from the YT videos. It’s not perfect, but it helps, and it gives me room to grow.

So, expect to see more entries here and on Buy Me a Coffee. I’ll be able to provide extra insight here and better explain things that may be lacking from the video content. It may also give me a chance to link directly to the video so the support material matches. It will probably take some trial and error to get things running smoothly, but with any luck that will happen quickly. So, here we go.