RAM type/speed for new system

[Guest D17S]

Unless the basics are understood, the rest becomes fuzzy and unintelligible. And here we are. So far, these underlying mechanisms had not been even integrated into the discussion. Relating the basics to global warming might finally explain this disconnect and the need for these technically difficult, then ad hominem argument responses as the inevitable requirement for additional detail occurs. However when this particular argument element begin to emerge in response to this normal learning dynamic, one need to consider even more carefully any detail elements contained therein.I believe this all might make sense, someday. But it will Never make sense starting from this mid-level. Global warming? Don't Even get me started there! Right now, Every one set your 'puters up to use its "Sleep" function. Never shut it off. If your electricity bill is pretty stable, you'll Actually be able to see the reduction, . . . and reduce your carbon footprint for us all.

[aerdt 0]

I am a little confused with all this talk about memory and overclocking. How does this benefit FSX?The last time I checked (and that was a few years ago that I used to do some overclocking), there was only a marginal, if not negligent performance increase from DDR to DDR2; so even if you tweak the mem speed up to 1000 MHz, you won't notice a significant, if any increase in performance in FSX.DDR = Lower latency (good), lower clock speeds (bad)DDR2 = Higher latency (bad), higher clock speeds (good)It kind of evens itself out in terms of "felt" performance, but most people think that DDR2-800 runs twice as fast as a DDR-400.Pat

[Guest Nick_N]

>Unless the basics are understood, the rest becomes fuzzy and unintelligible. And here we are. Well I

[Guest Nick_N]

NoelI sincerely apologize. You asked a question about memory product and I was sidetracked a bit with a dance around the dictionary. Although I thought I had answered earlier with a link to the Corsair 2x2GB perhaps you were asking if there was better.On the market right now, no. If I were going to run DDR3 2x2GB I would run the Corsair sticks I posted above which are available at Newegg. Unfortunately 2x2GB is still a bit behind for several reasons.. one of which, in my opinion, is the memory companies milking the system and they know what pro

[Guest D17S]

It's pretty much true that increased memory clock speeds can provide only "Felt," subjective improvements in FSX (given a normal budget!) This conversation has progressed on to attempting to understand why any improvements might be "felt" at all. The increased memory speed/FSX performance improvement premise - Should Not - be considered by normal builders. Use whatever ram that will allow you to run your FSB to generate a maximum O/C on your CPU. For instance, if you need to go to a 400Mhz FSB, buy DDR2-800 ram. Anything beyond that is science: You will be into the realm of the future of computer technology and will be provided a subjective improvement more in relationship to how much you spent for the stuff. And speaking of science, Consider the actual capability of our current FSB. At the quad pumped data rate on a 400Mhz FSB (400x4 or 1600), a FSB data event occurs every .07ns. Therefore, a data bit has the capacity to be uploaded onto this FSB every .07ns. Yes, 7 one-hundredths of a billionth of a second. At a 400Mhz FSB, a FSB data event occurs every .07ns. That's the FSB's data rate (FDR). These events are Not all used, but they are available. Think of it like a conveyor belt with troughs. The conveyor belt's speed is such that one of its troughs pass by a loading window every .07ns. These conveyor belt troughs may or may not get loaded. That is going to be a timing issue. But a ride on this FSB express is available at .07ns intervals. These numbers get amazing in hurry. On to Latency: Latency means "wait-time," right? Waiting for what? How about waiting to hop-on board the FSB express? Interesting premise anyway (might even be right!). For a moment let's continue with this proposed premise. (Guys remember, this is all imHo . . . but it sure makes sense). Remember the ram is running on its own Mhz cycle. It's been repeatedly suggested that the better the FSB and the ram cycles can be synchronized, the better performance will occur. Ok, let's go with that. That (might) mean that a ram latency of 45ns suggests the ram had to skip (wait-through) 641 asynchronous FSB data events before is was able to finally synchronize with the FSB to achieve a data transfer event. The data transfer finally occurred (the busses finally "synchronized") on the 642nd FSB data event.(Math? There are 642, .07ns intervals in 45ns) Adjusting either the FSB or the Ram speed to enable the ram to use more closely spaced "FSB data events" would reduce the ram's need to "wait for a synchronous moment" (aka "latency"). This could potentially increase performance. If Nehalam can deal with this issue successfully, we're in for a real treat. But it will have little to do with DDR3 ram, per-se. It will be about using more closely spaced Quickpath cycles with any user client (like ram). By this (proposed) analysis, changing the FSB speed (either up or down) is only helpful if it allows its clients (like ram) to utilize more closely spaced FSB generated data events. Maybe that's why increasing the FSB speed makes no difference. We have to understand this is not about speed like car. Every "thing" in the computer travels at the speed of light. It's all just electricity.This is - entirely - about timing. The trick is to allow the FSB and the ram to synchronize as often as possible. It appears that 45ns is about the record . . . but so far it has been only obtained on the basis of under-considered experiments. Even that old poky 200Mhz FSB provides a 1ns interval. If that 1ns interval syncs better with a particular ram speed interval, performance could actually increase at a 200Mhz FSB. For instance, if a 200Mhz FSB allowed some ram Mhz speed to sync-up to the FSB every 35 cycles, the latency would drop to 35ns. Is this fun or what!?

[Noel 4,484]

Thanks Nick for the clear recommendations.>If you find problems with the new system like you did with the>old and the replacement they sent, please verify the PSU. If>you see the same issue the only thing it can be at that point>is either a defective PSU, or,.. you damaged the processor. Wow, that goes against everything I've read, albeit casually read. I am running my QX within the max rated vCore, and actually, almost always I have run it at 4+GHz at much lower than max rated vCore. Plus, it has been kept very cool. I have to think it's not the processor, because of what I have said here. Here's what's actually happening now, with my ASUS supplied RMA replacement for my original P5E which was in service for only a couple of months. I am able to run my mushkin at 901Mhz comfortably, at 1.80v or on AUTO. FSB is only 375. It will not run at 400Mhz as my first P5E could. It has no trouble running stable and cool at 4.2Ghz. Beyond that it can heat up to 64C under high load (FSX for example). Prime95 runs with no errors. I have to think it's either the PSU, or a mainboard component as we originally suggested. But, who knows. Nick, I do have a question for you. Please forgive (read: check your ego at the door!) if this seems unwarranted. You have to remember I don't know you from Adam. You have a way of sounding VERY authoritative in your descriptions of the realities of hardware and computing in general, so from where do you get your extensive knowledge?Well, I'll have to see if I can get credit on an X48. Hey, what do you think about the Gigabyte x48 DDR3 board? ASUS is making me nervous now, with one dead board, an RMA'd board that *seems* crippled (ie, I haven't r/o the PSS per se), AND a second board was shipped to me at a discount, brand new, and it did one long and 4 short beeps, and while it eventually POST & booted, keep giving me BSOD's even on a fresh install of XP. It's being returned with the hope of getting credit on an X48 DDR3 board. Wish me luck. I may be stuck with another P5E, which if it works as the first one, will be awesome again. But, I'm getting to where I doubt that will occur.

[Guest Nick_N]

>I am running my QX within the max rated vCoreRemind me, what processor are you using?What is the max rated Vcore for your processor and where did you get this max rating?Quote:WARNING: Intel has not tested, and does not warranty, the operation of the processor beyond its specifications. Intel assumes no responsibility that the processor, including if used with altered clock frequencies and/or voltages, will be fit for any particular purpose.What is the Vcore you are running?With a 45nm CPU, as long as it is below 1.45 and the temp has remained below the initial thermal junction point, you should be fine.>from where do you get your extensive knowledgeI am not an authority. Although old school I did earn 2 PhD's, one in EE and one in TP and have been around the block more than once. I have in the past worked with Intel, AMD, IBM and others in partnership on systems for the military/aerospace industry and have been in the aerospace game since before Apollo. I retired early last year.>Hey, what do you think about the Gigabyte x48 DDR3 board? ASUS is making me nervous nowOut of all of the x48 offerings Asus is my preferred selection however I will pass this on

[Noel 4,484]

>What is the Vcore you are running?>>With a 45nm CPU, as long as it is below 1.45 and the temp has>remained below the initial thermal junction point, you should>be fine.OK, I'm fine then. My QX9650 runs at 4.2Ghz at 1.425v or less. I did run to over 4.4 at 1.45, but heat started to become an issue and besides, don't seem to need it at this point. I am usually running FSX around 4.14 to 4.2, or even 4.0. 4.0 runs totally stable at 1.35v, and stays very cool.>I am not an authority. Although old school I did earn 2 PhD's,>one in EE and one in TP and have been around the block more>than once. And, clearly you like to get into the detail on memory setup. Thanks for sharing Nick!>This is not seen on all boards and the common denominator has>not yet been 100% found, however thus far the 790i chipset>does not appear to exhibit this issue. Both the eVGA 790i and>Asus Striker II Extreme have been found to deal with high>clocks on quad/2x2GB very well although I can not vouch for>either because I have not worked with them.I am inclined to keep open the option of ATI CrossfireX, or not. I kinda doubt I would deviate to a 790i, but I really hadn't even considered it. >For more information, read Tony

[Guest Nick_N]

It was just a suggestion Noel since I just got the information yesterday. I have not had any issues at all here. Please do understand that with DDR3 1800 and 2x2GB @ 1T there is not a lot of room to play but the nice part is you are already clocked 200MHz over automatically so the Corsair sticks I posted are the best deal in town in the 4Gb solution right now unless there has been something released elsewhere I am not aware of.That is also why I advised against the CAS7 sticks @ 1600The nice thing about hitting 1800 is you really can not tell much difference when you pop 1T @ DDR3 1800 and tRD6 between CAS 8/7 and CAS 8/8.. and if you raise the subs a bit and add a little FSB for more CPU fun, that works too but @ 1T you wont be pushing it too far8-7-7-17 1T tRD6 @ 3.64Ghzhttp://forums.avsim.net/user_files/189179.jpg8-8-8-18 1T tRD6 @ 3.68http://forums.avsim.net/user_files/189184.jpgas you can see, because you are so far up the scale and because you are running 1T and MCH6, a simple 40MHz in CPU or 5MHz FSB has increased the performance and lowered latency. Therefore in this case relaxing the timing a tad is beneficial. If I had to for stability I could increase tRAS to 24 and see very little loss in this case. These are things you will learn as you play.. dont get discouraged and nit-picky like we would with DDR2 or even DDR3 13-1600 trying to nail that hot spot because at these bandwidths (1800 and up), although its really nice to be in that zone, its not DDR2 or even DDR3 13-1600 so you get more bang for very small FSB changes and your getting that bang on 2x2GB.If that was not the case Intel would not increase the native FSB on slugs.Understand?You do have a good slug to play with. That old P35 I am showing above is running a B3 QX 65nm so you should be in for some fun.I cant estimate or predict what you may be able to accomplish but none the less I know you wont be disappointed and you may need to take a bit of CPU reduction to hit the high note on the sticks but again, no worries.

[Guest Nick_N]

And one other thing Noel... dont hop right onto the settings Tony posted right away, work it out first and save that for last if you feel its not hitting the numbers right stableyou should be able to do 8-8-8-20/24 1T tRD6 stable with those sticks and do make sure they go in the black slots. Once you find your stable point I will give you some subs to try using MEMSET to make the changes and see if we can tickle another click or two in latency out of them. If stable in MEMSET over a good test period those values can be transferred to the BIOS so a MEMSET boot at Windows is no longer needed.tRD is MCH READ DELAY and I would not make that change in the BIOS at first.. I would set up the timing as listed and use MEMSET to make the Performance Level change to 6 in Windows for testing stability first.

[Guest ianrivaldosmith]

Bloomin eck! So you actually have to understand all of this these days to fly FSX at ultra high settings!?! Wow! Id have to read for a week to begin to understand this jibberish!Think I'll just stick with my quad clocked and DDR2 800mhz clocked to 960 or whatever its at :-p

[Noel 4,484]

>Understand?Sort of Nick . . .I became "sold" on CALWI when my first trial at 7.5ns or better with my OCZ/P5E seemed to translate to pretty much perfection. I think you are saying it's ok to go to 8.8ns provided you have the higher DRAM freq plus 1T, etc, but I guess overall latency is better, so it works. It looks like 1T may be a variable now always achievable, no?Nick, here is my fantasy: that I can use high end parts, run them close to spec, meet the low latency/high bandwidth, AND NOT HAVE IT BLOW UP! If this is really not realistic, I shouldn't even bother. It has already taken hours n hours of restoration because of my "playing." So, with the $549 kit of Corsair, and a P5E3 (hey, what about the Gigagbyte board? Seems all other users on newegg rated it very high versus the P5E3), will I realistically meet my goal?QX9650 w/ Retail HSF|8GB Muskin PC-6400|ASUS P5E|EVGA 8800GT @700|Seagate SATA 2 x 4|Seagate Cheetah 15K.x|XP Pro|Vista 64--soon to be installed

[Guest Nick_N]

Are you asking me for guarantees about overclocking?You know better than that. LOLAll I can do is post the information and perhaps help set up but you have to do the rest.If you don

[Guest D17S]

This is all just hobbiest entertainment. At best, the performance increase will be subjective, At worst, you'll damage your system. Builders should stick with DDR that will run at 2x your projected FSB. Users, don't even bother O/Cing. At best it will make NO difference. At worst, it will introduce an erratic instability. The most authoritatively presented information here is from a top-down perspective (in all respects). Results have been obtained from using the most expensive, cutting edge mobo/ram hardware available. We all would hope that stuff would preform better than current stuff and it does. The real disappointment has been that the improvement has proven to be so little. In this top-down approach, adjustments have been made to optimize these hardware combos, them explanations have been sorted to explain those results. Because "why" was not an initiating premise, when the question comes into play its encounter becomes extremely frustrating, especially when that question doesn't respond to the normal method determent. In the top-down approach, results come first, reasons for those results come after. This a Valid scientific method that is responsible for most of what we have today, but don't expect the fuzz to get any better.

[Noel 4,484]

>Are you asking me for guarantees about overclocking?Not really, I don't think. I am asking if I purchased higher end corsair DDR3 2x2gb, what sort of performance can I expect at spec voltage, or at least "safe" voltage, etc. I don't want to be on the bleeding edge, it takes way too much time to deal with poor outcomes.>If you don