re: PC Discussion - Gaming, Performance and Enthusiasts

I sold my H60 to Devious

I can fit the H80i but just need to clear the ram.

I also ordered a second SSD so that's two 512GB SSDs and one 3TB HDD in my HTPC.

It's going to be a nice little machine. I will probably upgrade the CPU from the 3770K later this year if it's worth it.

I am going to just run this one machine for a while. I have enough to build another PC but I am about to break down and sell it all on eBay.

I may go back to an ATX build one day when some new stuff comes out.

I just realized that DDR4 will be out later this year so I may return my order of that LP ram. I am just going to use one of the 8GB sets I have. I can fit it when I run air on my HTPC but I was wanting to add my H80i which is why I needed LP. I am just going to take the heat sinks off to get by for now until DDR4 drops later this year and I upgrade everything.

Regular GTX 650? 280x is roughly 3 times better or greater, depending on the game. We're talking a difference between 20fps and 70fps at the same settings at 1080p.

Also for a 650 Ti at 1080p, the 280x is a little more than double the performance.

Nice. Was just curious. How do you guys figure out how these different video cards made by different companies are better or not?

Passmark is probably useful for comparing cards with a large performance disparity like the 650 vs 280x. Generally though, I don't like using a score-based evaluation for GPUs, especially if gaming is the goal. There are cards that score lower in benchmarks but outperform the higher scorers in certain games. A good example is the GTX 770 vs. the 280x. They are basically equal cards in terms of average performance. The benchmarks on passmark show the 770 scoring nearly 20% higher than the 280x, but the 280x either matches or pulls ahead of the 770 in many instances, depending on resolution and how much the game depends on memory bandwidth. It ends up pretty even.

Another obvious inaccuracy in the chart shows dual-GPU cards scoring similarly or lower than their single-gpu counterparts, probably because Passmark didn't know how to properly deal with the 2 GPUs.

quote:

---

Nice. Was just curious. How do you guys figure out how these different video cards made by different companies are better or not?

---

With experience, one can make a general guess in performance based on the age of the architecture, its core specs (shaders, SPs, cuda, whatever), and memory bandwidth. Otherwise, a better (but still imperfect) barometer of performance is anandtech's convenient gaming benchmark comparison charts. Pick two cards and see how they compare in average framerate in a smattering of games at various resolutions and settings. Example: 770 vs 280x: LINK

Average framerate doesn't tell the whole story, but it contributes well enough for a reasonable decision.

As far as comparing the different AIBs (Gigabyte, MSI, ASUS, Sapphire, etc. etc.), you really just have to read reviews of those particular cards (reviews from legitimate tech sites, not Newegg). At this point I already have a general sense of how each AIB handles the fabrication and packaging of their video cards, and I choose based on my own requirements. For example, I always water cool my cards, so I need one with a reference design PCB and dual bios, and I tend to stick with companies that consistently use Hynix RAM (better overclocking) rather than sell cards with either Hynix or Elpida under the same SKU. I also choose companies that aren't over-the-top anal about what voids warranties and whether the warranty can be transferred.

It's cool, and I wasn't intending to brush away your advice as useless. I still use that Passmark site all the time for quick comparisons of CPUs because it shows multi- and single-thread scores and puts them through a multitude of tests. CPUs have all kinds of uses that it's more acceptable to use those types of benchmarks. Being able to see both single and multi thread scores at a glance gives me a general idea of how well it will pair up with a modern GPU in games.

I just saw your link while browsing reddit and was about to come here and say look at this queer pawning off his shitty equipment.

quote:

---

ILikeLSUToo

---

You been playing Borderlands lately? Like it?

quote:

---

You been playing Borderlands lately? Like it?

---

Yep, due to having no idea what to play. I started it a long time ago and for some reason got bored. Just giving it another shot now. It's not too bad, but kind of hard to go solo sometimes.