• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

MODS CLOSE THIS: NOOBS KILLED IT

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: Soulkeeper
guess i'm prob one of the only ones that notices high performance increases in the things i do from lower latencies.....
Click to expand...

Oh yeah? What do you do? How have you measured these performance increases and arrived at the conclusion that they're "high"?

And how can you justify even a 20-30% increase in performance (assuming that's what you mean by "high performance increase") for a nearly 100% increase in cost?
 
more in the ballpark of 10-20%
but yes only in a select few things with ddr and pc133 comparing 222 and 333
i havn't done anything recently which i could show benchmarks of
gaming would be a realm which this wouldn't apply tho
stuff like seti, pi, computational stuff dealing with many small memory chunks being accessed randomly/sparactically, and synthetic benchmarks are some examples
 
Here is a cool program i found awhile back and had the chance to run on many systems comparing the latency differences between different chipsets, cpus, and memory configs

i invite you to run this on your own system and compare 2-2-2 against 3-3-3 4-4-4 etc.
or higher options if you can't do them low latencies at the same mhz
you'll see the difference when you get out of the cache
I need to find an updated version tho as this one only tests up to 4MB
 
Originally posted by: Soulkeeper
more in the ballpark of 10-20%
but yes only in a select few things with ddr and pc133 comparing 222 and 333
i havn't done anything recently which i could show benchmarks of
gaming would be a realm which this wouldn't apply tho
stuff like seti, pi, computational stuff dealing with many small memory chunks being accessed randomly/sparactically, and synthetic benchmarks are some examples
Click to expand...

Oh, even better... 10-20% increase in performance for nearly a 100% increase in cost. :roll:

 
Originally posted by: Soulkeeper
i'm not speaking about cost
i'm speaking about the technology
Click to expand...


You called for a boycott so clearly you are talking about money.

Originally posted by: Soulkeeper
i'm done with this thread
Click to expand...


Brilliant, dozens of people tell you that you are wrong and offer detailed technical and anecdotal evidence yet rather than learn something from it you storm off.
 
Lower timings on DDR2 aren't necessarily a boon to performance. While my OCZ Gold sticks can run at 600 MHz with 3-3-3-8-1T timings, I get faster SuperPi times running them at 4-4-4-12-1T at 750 MHz. Frequency makes just as much impact on performance as the latency.
 
Originally posted by: Kwint Sommer
Originally posted by: Soulkeeper
i'm not speaking about cost
i'm speaking about the technology
Click to expand...


You called for a boycott so clearly you are talking about money.

Originally posted by: Soulkeeper
i'm done with this thread
Click to expand...


Brilliant, dozens of people tell you that you are wrong and offer detailed technical and anecdotal evidence yet rather than learn something from it you storm off.
Click to expand...



the only technical stuff i've gotten from the noobs that replied was telling me i'm stupid, ignorant, etc.
you kids got nothing to teach me, and that wasn't the intent of this thread anyways

all flamewar starters

 
Originally posted by: 996GT2
Lower timings on DDR2 aren't necessarily a boon to performance. While my OCZ Gold sticks can run at 600 MHz with 3-3-3-8-1T timings, I get faster SuperPi times running them at 4-4-4-12-1T at 750 MHz. Frequency makes just as much impact on performance as the latency.
Click to expand...
The point of this thread never was that 5-5-5-15@650 Mhz (DDR 1300) wasn't faster than 3-3-3-8@300 Mhz (DDR 600), it was that he would like to see lower timings. Of course, he doesn't believe that 4-4-4-10@400 Mhz, DDR 800, is exactly as fast as 2-2-2-5@200 Mhz. 'keeper, the people that told you that are right. As a matter of fact, 4-4-4-10 is slightly faster than 2-2-2-5@200, because of how much faster the RAM is running.

BTW, 'keeper, do you still have that 1.7 Ghz Williamette running at 2.8 Ghz? Also, I know you didn't/haven't asked, but with your PC3500 BH-5 and that IC7, have you considered just "upgrading" to a high-end Northwood? I know where you can get a 3.2C for $130. PM me if you're interested.
 
This was a very funny read. Soulkeeper, take a step back and read what the points people are trying to make. Ignore the insults and you might end up understanding where everyone is coming from.
 
lol, how can anybody take this thread seriously?
I would like to boycott toyota because they're not putting their F1 1000hp engines in my corolla --- the performance will be l337 h4x0r lol
 
Originally posted by: theprodigalrebel
I LOLed. Not at "MODS CLOSE THIS" but at "NOOBS KILLED IT".

I vote nomination for "Ownage of the Year".
Click to expand...

Self Pwnage is painful to watch.

(I speak from occasional experience 😱 😉)
 
Boycotting anything would only hurt the "small" companies that cater to us--Corsair, OCZ, GSkill, Geil, Patriot, PQI, etc. Do you think Samsumg would give a damn if .001% of its indirect customer volume was boycotted? Or maybe Elpida or Micron or Infineon or Hynix? Module manufacturers can only do so much to make RAM performance better and, believe me, they're doing what they can, as this would be a great competitive advantage. If you want RAM with tight timings then buy it at the insane extortion prices that are asked for it. Me, I'll just wait 'till the prices drop as they are wont to do.
 
Originally posted by: Mr Bob
I just wanted to bump it up for others to see lol
Click to expand...

thanks because this was a total waste of time. glad I didnt' miss it though, strangely.

very...informative in an a$$ backwards kind of way...
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top