Sign in with
Sign up | Sign in
Your question

Tight Timings vs High Clock Frequencies

Last response: in Memory
Share
April 2, 2006 6:19:10 AM

In this article, we take a close look at system memory performance in an effort to answer some essential technical questions. Can tight timings outperform high clock frequencies? How much of an effect does memory speed have on overall performance? Read on to find out.
April 2, 2006 3:32:51 PM

Well I really didnt expect the results to come out the way they did at all.

In my book its to close to be a noticeable advantage one way or the other, and I really felt that if anything the AMD CPU would take better advantage of tighter timings then the higher bus frequencey. Just goes to show you I guess.

But atleast this came out before I ordered my parts.
April 2, 2006 4:02:15 PM

http://forumz.tomshardware.com/hardware/modules.php?nam...

There's a list of my tests and results, I have done more but not updated it : p, when I have time (probably next weekend) and if I remember, Ill do it

Latencies sure help, but I think it's better to seek higher frequencies first and then try to tighten the latency
Related resources
April 2, 2006 5:22:42 PM

I know it's not as scientific or whatever, but a guy over at steampowered.com found a slight increase in FPS with tight timings over high freqencies in Counter-Strike: Source. This increase was only 5 FPS, but it's still something.

Source
April 3, 2006 7:17:53 AM

i have too better results in games with thight timings over high frec., dunno why :?
April 5, 2006 10:00:35 AM

Well, yet again the pros at THG manage to turn the tables and show you ultra-performance-geeks taht we noobies were right all along! :D 

No wonder these guys are respected... They have clearly concluded in this article that you DON'T need high-priced-high-end memory to play the latest games without stutter... that's the point I've been trying to make for so long now but all I got to this point are laughters!

And since most of their work is done in a near scientific environment I find no excuse to miscredit their results. Do you? :roll:
April 5, 2006 11:37:42 AM

do realise that when you increase frequency without changing the latency cycles, you're actually decreasing latency timings (if it takes 3 cycles to perform an operation at 200 MHz, if you increase frequency to 300 MHz the operation will take the same time as 2 cycles at 200 MHz. So, increasing frequency is a dual effect operation (more throughput, faster responsiveness).

The best being, of course, faster bus speed and tighter clock timings

There is a risk however, is that the RAM can't take the decreased timing and skip operations: if it can't go faster than 3 CL, setting it to 2 CL may have it skip every odd operation (which needs to be delayed) and actually increase timing to 4 CL!

Don't forget, too, that increasing frequency may lead to asynchronous HT/FSB and RAM clock - big performance hit.
April 5, 2006 1:29:32 PM

While HT may have tremendously diminished the impact of asynchronous operation, it doesn't annihilate it completely. And what I said is still valid for Intel owners - they still use a FSB with an external memory controller, in which case you can get a 20% performance hit.
April 5, 2006 2:45:48 PM

To think I almost bought some TCCD.
April 5, 2006 9:25:45 PM

Quote:
TCCD is still a good alternative if you mobo can't go above 3.2V and can find them at a decent price.

PDP Patriot 3200XBLK was the cheapest of all TCCD a while back.

My Biostar TForce does 3.3, and its fricken $69.
OCZ's Voltage Booster can add up to 4.0 also, provided it works with your motherboard.
What is the highest 2-2-2-5 TCCD will hit? I think 230ish right? And why does Winbond start then stop production of voltage-hungry chips so much? It's quite frustrating, but it does drive the resell value of my DIMM's up quite much =D
July 21, 2006 5:06:36 PM

That's all well and good, but where's the DDR2 freq. v. latency tests? DDR has been around for a long time, so, should I suspect that we'll see some DDR2 tests just in time for DDR3?
!