Well, I said I wouldn't, but I did buy a new system...

[brucets 94]

Thanks Nick----I've read Most of your posts and your XP guide several times. I need to dig thru your links again to find how to use O&O to get the right files on the outside of the disc and also the 3GB Switch as well(I may have found this one). I think I can stumble thru the XP set up. I am afraid to disable System Restore though. I'll get there eventually!Bruce

[Guest D17S]

I couldn't tell from the link. You can check it, though. Run HDtune (or HDtach). If you are seeing 70MB/s transfer rate, send it back! If you see a 100+MB/s transfer rate, that's the new one (I really expect you got sold the old one).If you still want to do the experiment, even the old one might help inform the conversation. Even the old style has a faster access rate. Access rate is supposed to be the FS holy grail and what argues to separate any Raptor from the rest of the pack. Technically, the argument makes sense, However, does it provide an even vaguely perceivable benefit? If so, how much . . . and if so, how much did it cost. Give it a try. That dual drive strategy is really just a housekeeping issue. If you keep your boot drive tidy 'n nice, there'll be no difference. If you defrag twice a year, do yourself a Big favor and install FS on that separate drive.

[Guest D17S]

Bump

[Guest Nick_N]

HDTach is a waste for other than CPU use and a look at the accessWhat does the header at the top of the GUI say?SEQUENTIAL READ SPEEDA/V is sequential read... games/OS/applicatiions are not and never will beFSX and any other game is NOT a sequential read file system, therefore that test is flawed for anything other than baseline examination and the result can lead someone to assume things that are not trueSo that means a professional RAID solution on HDTach is no faster than a single large platter drive?http://forums.avsim.net/user_files/189262.jpgHogwashWhen the right test is set up which represents the file system in use, and, the data chunks being called, then the benchmark result reflects the true performance of the application its being applied to http://forums.avsim.net/user_files/189263.jpgThe best way to look at true storage system performance is by using I/O Meter, set up sessions to record throughput during several FSX flights and then break the result down correctly. Cheap and fast benchmarks are only as accurate as the users ability to undertsand what the benchmark is doing, read the result and understand what they are looking at, and, their ability to understand the application with the goal of optimizing the storage system for its use.EDIT: And notice what happens at 64-128K when testing the FSX array when ATTO is set to look at average file calls based on the file system in FSX.

[Guest D17S]

Oh my. What an energetic response . . . but I really only wanted to find out which raptor he had. Honest. Sorry, really didn't mean to aggravate. Big breath. Sometimes that helps.

[Guest Nick_N]

SamAlthough true the GUI will typically display the drive serial/model the point was you can not run HDTach and call a drive type from the sequential read result. Running that test on a large modern platter drive can display a higher result than a Raptor, even though the Raptor is actually faster in the application.break time

[Guest D17S]

Apologies to the OP. If you'd like to run that experiment, we'd all be interested. Mod?

[brucets 94]

Sam -Well, it is ----- what it is, and it will stay where it is. I don't know enough to move parts around or replace them. I'll just have to make the best of this rig for 3-4 years and hope it will play FS11 decently for a year or more. I really enjoy reading your posts and Nick's as well; and this system is mostly based on what you guys, and others, have talked about. I'm sure lots of folks here have learned a great deal and we are very appreciative! I doubt I'll ever gain the knowledge and insight lots of AVSIM members have, but I have in deed learned a great deal from these forums! Thanks.Bruce

[Guest djt01]

However FSX average file size jumps to greater>than 512K, many being 1M average, therefore the larger STRIPE>becomes more important and because you are set up as a>partially sequential access, 256K works very well. >>Most 'games' are map-based.. meaning they do not load huge>files like FSX and therefore a smaller STRIPE would probably>be best for those. However some games do have massive file>size loads and those will take much greater advantage of a>256K STRIPE>Yes in theory the bigger stripe size would make sense with FSX, but going back and forth like I have with stripe sizes and FSX over the last two years from 8K up to 128K I did not see any difference in scenery loading. I have both O&O Defrag 8.5 and 10 and always use the

[Guest Nick_N]

http://forums.avsim.net/dcboard.php?az=sho..._id=39812&page=Review the 64-128-256K and up result