Low Latency, the lie...

Status
Not open for further replies.
ChaosBlizzard said:
You don't even know how memory works! If my system runs out of RAM it doesn't matter how much I have, if it's out, it's out! The system will then swap and I will take a performance hit...


No, it's not a physical 2.0Ghz machine, it's better.. Why have 2.0Ghz in one chip, when I can have that divided amongst two and handle more operations?

i know how memory works. you are focusing on the fact that you are maxed and when you upgrade it will be better. i am focusing on the fact that my slow laptop is running better than your dual processor machine memory wise. i doesnt matter if you have a 5ghz machine with 3GB of RAM...if it can't beat my 128mb Pc133 then it's worthless.

again i ask what multi-threading apps do you run? If you are running all single-threading apps on a dual processor machine there is no point. sandra gives you a boost in score because it can test it with multi-threading. But single-threading apps will only use one of your 1ghz processors.
 
You STILL do NOT understand. If the RAM is used up.. It will have to swap the data from the RAM back and forth into the HDD. This will add to the latency of the RAM bench. What part of that is hard to understand? Like I stated before... If the OS itself can use dual threading you get a performance increase.

"The first thing I want to make clear is: the HAL controls thread distribution, not the application. The HAL is like a police dispatcher, it receives hardware request and assigns it to hardware. This includes the CPUs. The next fact is that your CPU is NOT multitasking. It can only execute on thread at a time, period. One thread, no more. With this in mind, consider this, right now your Computer is running about 60-100 threads. Most of which are idle, however, some are active, and they must wait in line for your CPU time."

"Remember your HAL controls threads, and everything has threads, your OS, your apps, everything. So lets say you bust open Quake 2, which is single threaded. Will you see a performance increase with two CPUs over one? YES. Why? The HAL can dedicate a CPU to the Quake 2 thread, and run the OSs threads through the Other CPU. The Machine will run better. If your run a multi-threaded app, the HAL can grab the threads from that program and execute two threads at once. It will assign one to each CPU. This brings up a key point. If a p2 450 can only run one thread, and a dual p2 450 can run two, then why is there not a 100% performance increase? The two CPUs must talk to each other. This bus overhead causes a drop in the performance of the CPUÂ’s. So, in the end, you get an 80% increase."


Do you understand yet?!?!
 
4W4K3 said:
again i ask what multi-threading apps do you run? If you are running all single-threading apps on a dual processor machine there is no point. sandra gives you a boost in score because it can test it with multi-threading. But single-threading apps will only use one of your 1ghz processors.
Dude, your motherboard can also divide the work load. Your system doesn't have to rely on software...
 
Your acting as if my system "lags" while it runs Quake 2 lmao. What i'm saying is that i dont need multi-threading to have a fast computer. You are saying your computer is "faster" because it can handle 2x the threads? There is no set rate that the threads are assigned....it gives them as fast as the CPU can take them. and my computer will take them running 2.6ghz, while your computer will take them running 2*1000mhz.

Also like someone below said your coputer isn't totally dependant on CPU workoad. My RAM also will boost me becasue it runs 235mhz and can give my CPU information faster than your RAM will. It is also in Dual Channel which gives it a boost, something AMD XP setups are known for.

You should run a few other benchmarks as well...basing your computer on one single bench is inaccurate. You might find you get even better scores with another program. Try "super_pi" and see what kind of time you get. i'm interested in that program right now bcuz it's just got one test and is easily comparable.
 
4W4K3 said:
Your acting as if my system "lags" while it runs Quake 2 lmao. What i'm saying is that i dont need multi-threading to have a fast computer. You are saying your computer is "faster" because it can handle 2x the threads? There is no set rate that the threads are assigned....it gives them as fast as the CPU can take them. and my computer will take them running 2.6ghz, while your computer will take them running 2*1000mhz.

Also like someone below said your coputer isn't totally dependant on CPU workoad. My RAM also will boost me becasue it runs 235mhz and can give my CPU information faster than your RAM will. It is also in Dual Channel which gives it a boost, something AMD XP setups are known for.

You should run a few other benchmarks as well...basing your computer on one single bench is inaccurate. You might find you get even better scores with another program. Try "super_pi" and see what kind of time you get. i'm interested in that program right now bcuz it's just got one test and is easily comparable.

Actually, there is a set rate as to how things operate in a processor! Every processor has a slice of quartz that oscillates at a frequency that designates the chips timings!

What do you mean it doesn't make it faster!! Handling two threads at once doesn't increase performance? You had me fooled!

I didn't base my system on a single bench test.. Do you not see multiple scores on my signature? What good is that fast RAM if your CPU's can't take on a heavy workload?

Yes your computer will take ONE thread at 2.6Ghz.. Mine will take TWO threads at 2.0Ghz.. That's faster!
 
ChaosBlizzard said:
Actually, there is a set rate as to how things operate in a processor! Every processor has a slice of quartz that oscillates at a frequency that designates the chips timings!

What do you mean it doesn't make it faster!! Handling two processes at once doesn't increase performance? You had me fooled!

I didn't base my system on a single bench test.. Do you not see multiple scores on my signature? What good is that fast RAM if your CPU's can't take on a heavy workload?

Yes your computer will take ONE thread at 2.6Ghz.. Mine will take TWO threads at 2.0Ghz.. That's faster!

So you're saying my computer and you're computer take threads at the EXACT same rate? What is that rate? My processor is totally different than yours, i imagine the rate is different.

I'm thinking 1+1=2, but my single 2.6=2.6, mhz wise my computer is faster.

you're processing 2 threads at 2ghz (150*13.5?) and im processing 1 thread at 2.6 (235*11) i will finish a single thread faster than you would, but you would finish 2 threads faster than i would. so ultimately you get a better score in sandra then i do (in 1 of the tests) that i understand...but i would stick with my computer as it has MUCH better memory score and scores better in other benches i imagine. i dont base my entire system on my HD score either lol...as my HD doesn't play a major role (no gaming, nothing very intensive...saving pictures? lol)

You are using Sandra...that's it. i said use a different bench all together. Super_pi for example.

i'm not sure what you mean "can't handle a heavy workload". my computer runs lightning fast...i probably only use about 1.8ghz worth of it's speed...it's overkill for my use. There would be no point in me upgrading and getting a dual mobo if i had no purpose for it. which is why i keep asking you what you run that needs multi-threading? you can run sandra faster than me...and your OS (which is outdated and soon to be ditched by M$ all together) can use it. Other than that what programs do you have that will run multi? not everything is supported...
 
That's great.. So you have fast memory linked to a slower processor? I take it your master plan is to have memory that can do the work in place of the CPU? Yes, they are different processors.. While they may have different rates, THEY ARE SET! They don't fly up and down. So your using a bench testing tool I have never heard of.. Super_pi sounds like something I can buy for 99 cent on clearance..

Our processors also take different amount's of information at different SET rates.. P3's can do 6 IPC, I have two, so my system crunch's 12 IPC. AMD chips can do 9 IPC. More efficient than one P3.. However I do not have one P3, do I?

Also, why don't you think the HDD has anything to do with speed? In most system's thats a bottleneck. You do notice a difference if your HDDs can load data into the RAM faster.. If the data gets to the RAM faster.. The CPU's can work on it all the faster.

Also, Windows 2000 is not dated.. If you want to run an immature operating system such as XP, then by all means do so. However I have NOT crashed/froze once since I built this system. However I can't say that for Windows XP.

I go to ITT-Tech. They do NOT use Windows XP. They know better. I also have a friend that goes to DeVry. They are the same way!

Again, you must have a reading problem. The application DOES NOT have to support SMP to get a performance gain! I do NOT know how you came to that conclusion. If the OS and program each get CPU time and one or the other doesn't have to wait.. That is a major performance increase right there!
 
ChaosBlizzard said:
That's great.. So you have fast memory linked to a slower processor? I take it your master plan is to have memory that can do the work in place of the CPU? Yes, they are different processors.. While they may have different rates, THEY ARE SET! They don't fly up and down. So your using a bench testing tool I have never heard of.. Super_pi sounds like something I can buy for 99 cent at Shop-Rite..

Our processors also take different amount's of information at different SET rates.. P3's can do 6 IPC, I have two, so my system crunch's 12 IPC. AMD chips can do 9 IPC. More efficient than one P3.. However I do not have one P3, do I?

it's free...you really are mad aren't you? i'm not going to debate this if you refuse to go out of your comfort zone and try something new. super_pi is very popular with overclockers and it's somewhat new. (calculates pi) I would like to see if you get a better time running dual processors or if it doesn't effect it.

Since when is a 2.6ghz processor slow? it's faster than both of your combined mhz wise. My system is running 1:1 235mhz DC...my processor runs at the SAME rate as my RAM...so i don't know what you are meaning by "faster ram and slower processor"

Yours can handle 12 and mine does 9....but you forget AMD is more efficient than Intel processing wise, AND my CPU is faster mhz wise than yours. Hence why i got a better score in 1 sandra test. You beat me with the other test because indeed it benefits from the dual processors.

i think you are taking this personally or something, i'm just trying to learn. If your computer only beats mine in 1 test in Sandra...and you base that as "my coputer is faster than yours" then that's not very accurate. That's why i suggest you run a different program and see if it still is faster.
 
ChaosBlizzard said:
I go to ITT-Tech. They do NOT use Windows XP. They know better. I also have a friend that goes to DeVry. They are the same way!

Know a guy named Caleb? he goes to Devry in Texas. 2000 is indeed old and Microsoft will stop updating it in a few months, get ready for system holes and no patches my friend. 2000 has many bugs that will go unfixed...XP has problems yes, but it is fully supported and updated constintely. I would prefer an updated OS over one that will be left to the dogs. Might i add you said you do not overclock, of course you do not crash. I overclock the snot out of my machine and reboots are expected. The main issue will be viruses/hacking..i have been on my home network for i think 7 yrs. now and never once been "attacked", that's all i care about since i kow my system is going to crash while i test it's overclocking ability. i guarantee you your 2000 based machine will crash if you set it at 200*10 trying to overclock it. and if i try 300*10 mine will crash...OS has nothing to do with overclocking/hardware overclocking.
 
Your system is not faster.. There you go again, basing everything off MHz. There are other things in this world that affect the speed of a CPU! Such as but not limited to cache, IPC, bus, core clock, programming.

"Yours can handle 12 and mine does 9....but you forget AMD is more efficient than Intel processing wise" - Where do you think they go the efficiency from?!?! I will give you a sudden clue.. It has to do with IPC!

Your right, those test's benefit from SMP. In much the same manor as NORMAL SYSTEM OPERATION. The system is always processing two threads.. Not once will it only take on one! That does increase performance. No matter how you look at it SMP is more efficient.

Don't believe me? Go on the internet and read some more.

I am not taking this personally. I just fail to realize how someone who refuses to read can write so much and not know how it works. I have been building computer for 6 years. I now go to a tech school.
 
Status
Not open for further replies.
Back
Top Bottom