Newegg now has the e8500 for $199, and the e8400 for $179.

That extra half a multiplier (9.5 vs. 9.0) should give me a greater assurance of being able to overclock (e.g., will only have to take FSB up to ~421 MHz vs. ~444 to reach 4 GHz). Does that in itself justify the $20 difference?

"Bragging rights" for the 8500 is kind of silly for me--i.e., not an issue. But I am tempted to "go the distance" since this is my last processor until Nehalem and Mobo are more mainstream (1.5 years?). For example, it would be great of the 8500 gives me an extra 200 MHz of OC headrooom to take the processor to 4.2 GHz a year from now if needed.

Yet out of principle I don't want to drop $20 if it really is only to feel good.

Thoughts?

TBH I wouldn't bother. How many times will even the E8400 run at 100%? Most time it will be idle. Most of the time when it's not idle, it will be limited by the HDD or video card, depending on what you're doing. You might as well invest the $20 in beer

Maybe you should consider a quad, like Q6600, too. What kind of programs are you running?

BTW, looking at your config, the first thing I'd upgrade is the HDD, not the CPU. Those older WD disks can't really compete with the newer stuff like the WD6400AAKS.

Hi aevm:

Thanks for the help. Business/office/productivity during the day, and BF2/2142, etc., in the evening. I am going to do a massive re-rip of my entire CD collection into a higher resolution soon (lossless this time!), but no editing or anything (which is why I am really not considering the q6600).

Funny you mention the 6400AAKS--the plan is to buy 2 of them for a Matrix RAID configuration in the coming weeks/months. Glad to see them coming down in price as I procrastinate! Since the price for 2 of them is identical to the e8400, would you recommend doing that over the e8400? The new hard drives will speed game loading, but nothing else in terms of gaming benefits, right? I am also waiting on them to do my music collection re-rip.

Then onto a 4850, a new monitor, then another 4850...

Even game loading won't benefit that much for RAID. You would need to run several hundred mb files to get some gains.

The applications you mentioned will work just fine on a dual-core indeed.

Yes, I'd prefer the new HDD over the new CPU if I can't have both.

The ripping part will definitely benefit from a faster hard disk. Apart from how fast you finish the job, there's also the fact that a slower HDD may cause defects in the WAV files if it can't keep up and the software isn't smart enough to synchronize properly.

Just curious, are you going to compress the files, or just keep uncompressed WAV files on disk? With WAV, you get no loss at all, and the CPU doesn't matter at all, but they take about 10MB/minute.

With MP3 at 256kbps you cut that to a quarter, and I think most people wouldn't hear a difference. I certainly couldn't, and I tried hard, using pretty expensive toys (Cowon iAudio 7 + Shure e4C headphones, X-Fi XtremeMusic + Z-5500).

I know there are lossless compression algorithms out there, but TBH I haven't tried them. For one thing, they are less portable and I want to choose my mp3 player by sound quality, not by the formats it supports. For another, I understand the files are still large after that kind of compression.

In fact, if we're talking about RAID 0 here, it's probably best avoided on a machine that's also used for work. RAID 0 increases the risk of losing data.

If we're talking about RAID 1, that's good for a work machine, because it mirrors the data and increases its safety, but it's overkill. It also loses the capacity of the 2nd disk, i.e. you pay for two 640GB disk and have a total of 640GB of space, not 1280.

Hi guys:

Thanks galta and aevm.

I had been teetering on the RAID anyways. May just go with a single 6400, then create a smaller "music" partition and RAID1 that with my existing slow 250 GB HD, then use an outboard 160 GB as a system backup, sans the music.

I am an audiophile and had ripped my collection at 320 kb/s with iTunes, assuming that was enough. It is not. I can hear a slight difference via the iPod (plus Headroom amp and Etymotic buds). But when I burn a redbook CD from the ripped files, 320 is night and day horrible. I am going to go FLAC/Apple lossless this time around. Much of my collection is based on CDRs, and some CDRs from 1999 are starting to go bad. I want to have everything on HD, (with HD backed up), in addition to hard copy.

So I guess the short answer is go 8400, and bank the $20 for use on the 6400AAKS...

The plan was to Matrix RAID, with a small RAID0 partition for OS and games, then RAID1 for the music, pics., docs., etc. I would then use a 160 GB to back up the system files and misc documents.

Although I am told that arrangement would work, the performance benefits really would not justify the second 6400. Let alone that I could use the existing 250 as a backup.

I appreciate your comments!

You're very welcome. And congratulations, you've got better ears than mine

Nah, probably not. I just know what to listen for.

So unless anyone recommends otherwise, I will just snag that e8400 and call it a day.

For a new build, I would go with the E8500 over the E8400. Percentagewise, the price difference is fair.

However, since you are already overclocked to E8400 speeds, you will not notice the difference without overclocking. Even then, I doubt that you are running at 100% under heavy load. Turn on the performance monitor and check that out. A $200 expenditure for no real difference is not worth it. I would wait for nehalem, which should be out by the end of the year. How does a $300 quad with hyperthreading and 25% more clock for clock performance sound? I'm waiting.

If you want more performance, splurge on a velociraptor. You won't have to mess with raid.

The value of raid-1 for protecting data is that you can recover from a hard drive failure quickly.
It is for servers that can't afford any down time.
Recovery from a hard drive failure is just moments.
Fortunately hard drives do not fail often.
Raid-1 does not protect you from other types of losses such as viruses,
software errors, operator error, or fire...etc.
For that, you need EXTERNAL backup.
If you have external backup, and can afford some recovery time, then you don't need raid-1.

Sorry but I have to completely disagree with that. I had an e6750 overclocked to 3.4ghz before I upgraded to my e8500. Even with the e8500 at stock speeds (3.16ghz) it completely blew the e6750 out of the water!

I was totally shocked so ran lots of benches and double checked everything but the truth is the e8000 series are far superior to the e6 series.

It really depends on what your going to do with it. Don't go for the e8500 if your doing mostly office stuff and then the occasional gaming sesh in the evening as your not going to benefit that much from it. but, saying that it's only $20 dollars! What the hell go for it... these things overclock like crazy and with the e8500 you can easy hit 4ghz without even trying....Good Times

Very very weird. The E8500 at 3.16 and the E6750 at 3.4 should be extremely close, except in SSE4. You didn't change anything else at all, just the CPU?

Here are some benchies showing E8400 vs E6850, both at 3 GHz. I'm assuming e6750 at 3GHz is exactly like E6850 at 3GHz, btw.
http://en.expreview.com/2007/11/30 [...] 00/?page=5

Based on those numbers, the E6750 at a 7% higher clock (3.4 vs 3.16) should in fact be faster in a lot of tests...

What kind of benchmarks did you run? Were they heavy on SSE4 or on floating point crunching?

I just ran the usual really...

Pi, 3DMark 05/06/Vantage, Half-Life 2 and from my own experience of video rendering using Sony Vegas Movie Studio.

Everything else was exactly the same, although I did have to update the BIOS for my Asus Maximus Formula to identify the chip correctly but I remember that I wanted to do like for like so FSB, memory, GPU were all kept at the same stock speeds.

Oh yeah! and I remember Crysis having a BIG improvement in average FPS.... although sadly it still couldn't force me to play a crap game ;-)

EDIT: Interesting, just read that article and it does show them as being pretty close. hhhmmm but something is telling me that review is not quite right... there should be a bigger difference than that. Most of the time is just a few percent and it's only occasionally it rises to being higher. Will have to dig around for similar reviews to see if they match.