• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Building a new computer

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Impact on gaming? Okay, maybe. But in no way will it be a minimal impact on overall system functionality. Everything that uses any real amount of memory done on a system in single channel mode can be as low as half as fast as it can be in dual channel. That includes everything from every day web browsing, to running just about any application (including games, which use memory to store calculations done by the processor). Think about it: you are literally running your memory at half of it's rated DDR (double data rate) speed. That's a huge difference, especially since that it's usually the same price for a single 8GB stick at 1600MHz (example) as it is for a 2x4GB kit at the same speed and timings. And the speed difference gets bigger the faster the ram that you get.

As a general rule, use the number of channels as your platform supports. For the least amount of strain on the IMC in the CPU, it's best to use exactly the amount of memory DIMMs as the number of channels available (most are dual, higher end platforms can be triple and quad).
 
It doesn't tackle dual channel vs single, but here is memory scaling on Haswell, DDR3 1333 to 3000, CAS7 to 12:

http://www.anandtech.com/show/7364/memory-scaling-on-haswell

25b%206950%20Bioshock%20Infinite%20Minimum_575px.png

26b%206950%20Tomb%20Raider%20Minimum_575px.png

4%20WinRAR%204.2_575px.png

8%20x264%20Pass1_575px.png


Basically, in games, there is no difference at all between DDR3 1333 and DDR3 3000. Compression shows good scaling, and encoding shows slight scaling.

At this point it's largely academic, because you may as well get dual channel if your board supports it (unless your board only has 2 slots), but I can do some benches on my i3 and i5 systems and see how much single channel hurts performance, if you'd like.
 
If you want to, go for it. The first one I would look at is system startup time. I have a feeling that it will be noticeably affected. Next, I would look at things like opening/closing the web browser, opening and closing multiple tabs in the web browser, and going into your file system and creating a simple text file, closing the Explorer window, and then going back into your file system to retrieve and delete it.

These are the things that I feel will be both done and affected the most.

If you feel like running the benchmarks yourself and proving me wrong, by all means. I will admit that both 1: I had a laptop that shipped with 1GB of single channel memory, and installed a second gig. I know that 2GB of memory is way low by today's standards, and it was an old system a long time ago, but I could swear I felt a difference. and 2: I very well may have only 'felt' the difference by knowing that I had dual channel, and no real quantifiable evidence will suggest that it's the case. If you don't want to test it, that's fine too. I don't really feel like messing with my system for something so trivial myself, so I wouldn't blame you.
 
If you want to, go for it. The first one I would look at is system startup time. I have a feeling that it will be noticeably affected. Next, I would look at things like opening/closing the web browser, opening and closing multiple tabs in the web browser, and going into your file system and creating a simple text file, closing the Explorer window, and then going back into your file system to retrieve and delete it.

These are the things that I feel will be both done and affected the most.

If you feel like running the benchmarks yourself and proving me wrong, by all means. I will admit that both 1: I had a laptop that shipped with 1GB of single channel memory, and installed a second gig. I know that 2GB of memory is way low by today's standards, and it was an old system a long time ago, but I could swear I felt a difference. and 2: I very well may have only 'felt' the difference by knowing that I had dual channel, and no real quantifiable evidence will suggest that it's the case. If you don't want to test it, that's fine too. I don't really feel like messing with my system for something so trivial myself, so I wouldn't blame you.

My bet is that that larger part of the speedup you experienced was from the increased amount of RAM, rather than the bandwidth, because those are all very heavily disk I/O limited operations. Windows has a feature called "superfetch", which will automatically fill your RAM with preloaded programs so they don't need to be fetched from the disk when you want them, providing a tremendous speedup. Basically, a "smart" RAMdisk that doesn't require provisioning. The more RAM you have, the more that can be loaded into RAM, and kept in RAM while other programs are running.

Those specific tasks would be difficult to quantify as there's nothing you can just download and run to get numbers with, and probably wouldn't be comparable across all modern systems either. For instance, AMD's memory controllers are significantly less efficient than Intel's, providing something on the order of maybe 30-40% less bandwidth with the same memory speeds, so single channel on a modern Intel platform would have rather close to the bandwidth available an AMD system has in dual channel, and less latency. Then you have i3's, which are basically half of an i7, and should theoretically need only half of the bandwidth.

I'll see if I can write a script to record how long it takes to do these things.
 
I know there is an application that measures boot time, it may even be called boot time or something to that effect. Aside from that, I was thinking a simple stopwatch would work for most of those tests. Up to you as to how you want to do it, if you want to do it. So long as it's accurate and can be repeated.
 
My bet is that that larger part of the speedup you experienced was from the increased amount of RAM, rather than the bandwidth, because those are all very heavily disk I/O limited operations. Windows has a feature called "superfetch", which will automatically fill your RAM with preloaded programs so they don't need to be fetched from the disk when you want them, providing a tremendous speedup. Basically, a "smart" RAMdisk that doesn't require provisioning. The more RAM you have, the more that can be loaded into RAM, and kept in RAM while other programs are running.

Those specific tasks would be difficult to quantify as there's nothing you can just download and run to get numbers with, and probably wouldn't be comparable across all modern systems either. For instance, AMD's memory controllers are significantly less efficient than Intel's, providing something on the order of maybe 30-40% less bandwidth with the same memory speeds, so single channel on a modern Intel platform would have rather close to the bandwidth available an AMD system has in dual channel, and less latency. Then you have i3's, which are basically half of an i7, and should theoretically need only half of the bandwidth.

I'll see if I can write a script to record how long it takes to do these things.

You won't see any performance difference from increasing the amount of RAM unless you're completely filling the available RAM and causing the system to cache to the HDD/SSD.
 
You won't see any performance difference from increasing the amount of RAM unless you're completely filling the available RAM and causing the system to cache to the HDD/SSD.

You'll see programs load faster (especially those less frequently used) because Windows will cache more them in RAM, and flush fewer from RAM when you open other programs.
 
You'll see programs load faster (especially those less frequently used) because Windows will cache more them in RAM, and flush fewer from RAM when you open other programs.

If your programs are all cached on boot-up, then why does an SSD make commonly used programs open faster?
Right, because they aren't loaded to the RAM until they're opened.
 
If your programs are all cached on boot-up, then why does an SSD make commonly used programs open faster?
Right, because they aren't loaded to the RAM until they're opened.

I'm sorry, but you're wrong.


http://www.osnews.com/story/21471/SuperFetch_How_it_Works_Myths

SuperFetch' second goal is to make applications launch faster. SuperFetch does this by pre-loading your most often used applications in your main memory, based on not only usage patterns, but also on when you use them. For instance, if you have the same routine every morning (Chrome - Mail - Miranda - blu), SuperFetch will pre-load these into memory in the morning. If your evening routine is different (for instance, it includes Word, Excel, and Super Awesome Garden Designer), SuperFetch will adapt, and load those in memory instead during the evening.

When people look at the Task Manager, and they see the figure for "Cached" compared to the number of "Free", people assume that only very little of their memory is available for the applications they are about to launch. What they forget is that the Cache filled by SuperFetch and the standard caching mechanism runs on a lower priority; in other words, memory requests by applications will always supersede SuperFetch.

In other words, whatever you see in the "Cached" figure is actually accessible to applications.

And this brings us to the question of what to do with RAM. I have 4GB of main memory in my main desktop machine, and I would find it a total waste if the operating system did not use it to make my computing experience smoother. Isn't that why I got 4GB of top-quality RAM in the first place? To make my machine faster?

This is exactly what SuperFetch does. It's an intelligent mechanism that uses the RAM in a machine to its fullest potential to make computing a smoother experience. The fact that SuperFetch (and its related technology, ReadyBoost) actually works, has already been confirmed by Tom's Hardware. The key here is that the more RAM you have, the bigger the benefit SuperFetch delivers; according to Tom's Hardware, Vista's sweet spot was about 2GB of RAM, but even at 1GB they noticed a positive difference.


http://en.wikipedia.org/wiki/Windows_Vista_I/O_technologies

SuperFetch is a technology that pre-loads commonly used applications into memory to reduce their load times. It is based on the "prefetcher" function in Windows XP.[9] SuperFetch attempts to load commonly used libraries and application components into memory before they are required. It does so by continually analyzing application behavior and usage patterns, e.g. what applications are typically used in the morning after logon.[10] The cache memory is marked with low priority, meaning that if another process needs the memory, it will be given up.


However, this is also important to know:

http://www.thewindowsclub.com/disable-superfetch-prefetch-ssd

SuperFetch and Prefetch are Windows Storage Management technologies that provide fast access to data on traditional hard drives. On Solid State Drives they result in unnecessary write operations.

Windows 7 and Windows 8 therefore by default will automatically disable SuperFetch and Prefetch, once it detects an SSD on your system.
 
So you're citing a feature that's disabled by default for the vast majority of users on this forum...

Still doesn't answer why having an SSD still makes commonly used programs open faster.
Memory is vastly faster than an SSD, but a system with an HDD would load Chrome way slower than one with an SSD.
 
It was relevant in this context:

Dlaw said:
I will admit that both 1: I had a laptop that shipped with 1GB of single channel memory, and installed a second gig. I know that 2GB of memory is way low by today's standards, and it was an old system a long time ago, but I could swear I felt a difference.

Windows still does caching, it's just that it does it in a different way with SSDs.

Anyway,I'll get back to you guys tomorrow night with my single vs dual channel benches, but I'm just putting it out there that dropping from 8GB to 4GB of RAM *will* make a difference.
 
Back