Create an account or sign in to comment
Because all of the above. I appreciate your concern, but there is nothing I can do about it. There's a glaring problem in your speed benchmarks.
You can say "probably" all you want, but you don't know unless you actually do it. That is what benchmarking is all about. It's not a 'glaring problem'. Actually, even slight manufacturing differences would affect the performance more than the amount of RAM. To my knowledge, RAM performance isn't affected by usage as long as you're not swapping. OP said that the amount of RAM wasn't a bottleneck.
You should test PlanetSide 2 if possible of course. Make sure that you have the game is as CPU limited as possible though. Here are my findings for those interested: Well RAM speed is a different test to amount of sticks.
After all it's both the memory bandwidth single vs dual channel and memory speed contribute to the memory throughput. The ideal situation in real world benchmarks is to test current popular games, because that's actual real world.
It may be a fun conversation to talk about PlanetSide 3 whatever that is , but it's a waste of time compared to testing the actual games that people would be playing since that's effectively all that matters.
The idea of a real world benchmark is to create a real world expectation of your performance gain, not simply to use games that make it easy to benchmark or even games using a popular engine. A real world benchmark is ideally the top most popular games and apps for platform X and broken down into genres, like FPS, simulation, RTS and so on and so forth.
You'll see more difference testing 2x2GB rather than 2x8GB simply because you're accessing less of the RAM to run your tests, but the first test in Spodemark is a good test of memory bandwidth with a fast enough CPU. The main point of my post was to prove how it doesn't matter in non-APU gaming machines.
So many builds recommend dual channel "just because" when the only use of the machine would be gaming. I'd say it's still best practice because if you fill a slot with a single module, you can't guarantee that the same module from the same manufacturer in future won't be subtly different enough to not play together nicely. Kits of RAM are tested and warrantied together, and RAM being as cheap as it is at the moment, it makes sense to buy it together. Good info duckne, I always wondered if there was a big enough difference between dual vs single to suggest going for dual even at a higher cost.
I would try it with a memory intensive game, maybe skyrim with mods? You sir have ran the test I have not had the resources to run. I thank you good sir, we need more scientists like you. And we need less of the armchair scientists who criticize his methods but refuse to do any of the work themselves. One "advantage" you have left out, is it does actually look better and more balanced in my eyes to have 2 RAM slots populated vs.
This may seem like a petty quibble, but there's a reason with the Sabertooth is one of the top 5 most popular motherboards on Amazon and it's not cause of its great price Those were the days! I think the reason is that people want to have the flexibility to re-use old DDR chips and when you have only have 2 slots you don't have much flexibility. I would say it's more of consumer ignorance than anything as the "Sabertooth" branding is pretty successful.
I'm sorry but you cannot say that looks have no bearing on PC purchasing decisions. People spend a lot of money on cases with windowed side panels, LED lights, custom cable sleeving and so on. I'm not saying that this is the no. It's just as valid as your second argument that "if one fails etc". Neither has any affect on the performance of the build, they're both "quality of life" advantages for the consumer. People spend waste a lot of money on cases with windowed side panels, LED lights, custom cable sleeving and so on.
I'm sorry but I must have phrased my answer above wrongly, I do apologise for the misunderstanding. I'm not saying that looks have no bearing on purchasing decisions, I do wholeheartedly agree with you on that matter. What I'm disagreeing on is the contention that:. Looks are largely a personal opinion. To you it looks better. But I was stating an extreme example whereby someone recommends 4x2GB simply because "it looks better".
One could even claim that he "likes the look of just one stick because it is minimalist". Personally, I feel that for recommendations, we should not get our own personal tastes into the matter. This was my beef with your statement. True, it's a very personal preference which is why I put "in my eyes". Perhaps it's not though, and I'm generalizing too much! One advantage to having one stick by the way is it is easier to diagnose memory issues and there's less to go wrong.
That's the main reason why 4x2gb is a bad idea I think, is that with 4 different sticks and 4 different slots, it takes awhile to narrow it down to the faulty component. For most people 8 GB is enough, chances are you probably know if you need more than 8 if you're going for a 2x8 setup.
Not having enough RAM will be a much bigger hit to real-world performance than not having your sticks in dual-channel.
Right now Newegg is giving away a free 8 GB stick with certain motherboards. You'll notice very little real-world difference if you opt to take advantage of that deal than if you bought another stick to put it in dual-channel or grabbed a 2 x 4 kit instead. Hmm, time to figure out what I was looking at then.
Dual-channel does increase the memory bandwidth though so to some extent that benchmarks are still relevant to the DDR vs DDR discussion. Basically that the test was completely void because OP didn't test 1x8 and 2x4 but instead tested 1x8 and 2x8. Dual channel definitely doesn't really do anything.
In my view it should all come down to price since the difference in some benchmarks are low anyways. Dual channel is more because it very likely costs more tp make two 4GB sticks than one 8GB. The price can be justified because of the doubled bandwidth. If you can utilize that increase depends on what you do of course. Which isn't really a big deal at all So unless it offers a huge speed increase with things like video editing I dont think its worth the price much of the time.
Submit a new text post. Rules Be respectful to others. No piracy or grey-market software keys. Limit posts to one every 24 hours. Do not use all-caps post titles. Link shorteners are not permitted. No intentionally harmful or joke advice. No offering rewards to solicit help. Pick, Assemble and Install: Welcome to Reddit, the front page of the internet. Become a Redditor and subscribe to one of thousands of communities.
As it stands right now, the common consensus for single vs dual channel is 2xRAM Advantage: Want to add to the discussion?
All you haters should use some logic for more productive suggestions. This is a fresh wind and it should be applauded. Nothing but respect for the time and effort you have put in this post. IMO, you're time is better spent not repeating other people's work.
I did it once each, but on hindsight, should have done averages. I did x video encoding as well, there was minimal to no difference. If you know of a reason it would make a difference I'd be fascinated to know about it. Actually we do know. Here's how paging works.
I'll put that in my list then if I have the time. Get it to test and get it to play. Wish I had some spare cash to test modern hardware. This sounds like fun. You have to have pairs of DIMM's for it to work. Even in boards that support dual channel most all of the DDR-4 boards that I know of do if you only use one stick then it can only operate in single channel mode. There are also triple channel DDR-3 X58 chipset boards that require sets of three and quad channel that use sets of four.
I hope that explains it a little better. Intel Core i7 K. ASRock Z77 Extreme 6. Yes, it's just a simple multiplier The more important question is whether it actually matters If you are not even using the bandwidth provided by Single-Channel, then having twice as much will do nothing at all. Most things do not scale beyond dual channel, but pretty much all modern systems support at least dual channel. Whether it matters compared to single channel frankly is not widely tested, because all systems are capable of dual channel so there's no reason to test if you need it.
Whether you need it or not, you have it. It's faster, depending on what you plan on doing with it. What exactly are you looking at buying? In gaming, it helps in scenarios where you are cpu limited and some titles that really seem to like a lot of memory bandwidth, but with most games that are gpu limited, you won't notice any difference. Yes, there's nothing actually different about the RAM in single-channel or dual-channel kits.
You need to be a member in order to leave a comment. Sign up for a new account in our community. Already have an account? By Edward78 Started 5 minutes ago Posted in Troubleshooting. By aandril Started 10 minutes ago Posted in Photography and Videography. Or sign in with one of these services Sign in with Facebook. Sign in with Twitter. Sign in with Microsoft.