WD360ADFD HDTach Results

Not too many people own these drives and many are curious about them. I have owned mine for about 3 months now, no problems at all, it's as quiet as any seagate or western digital I have ever owned, but it is fast.
I would say, if you don't need a lot of room, or are just using this as a boot/operating system drive, this would work good. I have never been able to fill up more then 20GB on my main machine, I leave all my music/movies/etc on my server.



It seems to be a consistent 10MB/s throughput slower then the 16m/b cache 74GB'er but it has about .5-1ms Random Access Time faster. Same average Burst Speed.
The 36 also has a consistent graph...I have seen many WD740ADFD's that have a drop or two, mine has none. I ran this test multiple times and got the exact same data everytime.




This is about $50 cheaper then the 74GB version. If you only need 36GB on your main disk...IMHO go for this one.

It seems to be a consistent 10MB/s throughput slower then the 16m/b cache 74GB'er

Click to expand...

I have two and don't find that true at all. Single RAID0

The 36 also has a consistent graph...I have seen many WD740ADFD's that have a drop or two, mine has none.

Click to expand...

Has nothing to do with the drive. Just means something probably tried to access the drive while running the benchmark.

If you only need 36GB on your main disk...IMHO go for this one.

Click to expand...

No doubt. Performance is the same for all the new Raptors.

lol...Can we swap drives?
jk
So I guess your drives are as fast as the 74GB version. I wish mine were as fast, at least they're not too far off.

Question, in your opinion, is it faster to Raid-0 two Raptors, or use one raptor for Windows and the other raptor for games?

For reference, here's my 2 WD740ADFD's in RAID0:

So I guess your drives are as fast as the 74GB version. I wish mine were as fast, at least they're not too far off.

Click to expand...

When I got my second ADFD drive and it was only getting 71MB/s I called WD and told them it was not acceptable when my other drive did 78MB/s. They RMA'd it and my new did 76.4MB/s. Petty as it may seem, I didn't pay a premium for the drive to get "average" performance. If you're getting in the 60's I'd be calling WD today for a replacement, unless you can attribute it to a poor SATA controller/drivers.

Question, in your opinion, is it faster to Raid-0 two Raptors, or use one raptor for Windows and the other raptor for games?

Click to expand...

RAID0.

Megabytes per second has very little to do with actual performance. If you'd been getting 13 ms seeks it's another matter, but returning a drive for falling nine percent short of your personal STR goal seems petty at best. Did you do any benchmarking, or just assume that it was broken?

Did you do any benchmarking, or just assume that it was broken?

Click to expand...

Well yeah, HD Tach, ATTO, HD Tune. That's how I determined it wasn't performing up to spec. Far as I'm concerned, I didn't sell my GD Raptors to buy ADFD Raptors with GD STR performance.

Megabytes per second has very little to do with actual performance.

Click to expand...

I agree, but WD claimed a 12MB/s increase in STR's between the two, I should get something close to it for my money.

HD Tach et al aren't benchmarks in the performance sense - they'll tell you if you have broken hardware, but they won't tell you how well it performs doing high-level operations. So if you had gotten half the STR or twice the seek times you expected, it would've been a good idea to check with another controller or another machine, and RMA if your results were confirmed.

But to run several benchmarks that do essentially the same thing and assume that 10% STR leads to 10% performance decrease isn't a good way to do it. Play some games, whip out the stopwatch, and get some numbers. HD Tach is completely synthetic - you wouldn't expect a hard drive with doubled transfer rates to perform precisely twice as well, would you?

Wait, look who I'm talking to. Oh well.

Wait, look who I'm talking to. Oh well.

Click to expand...

I forgot I'm on the [T]ard forum where people make assumptions and put words into people's mouths. Simply put, I don't want a 10% drive variance in STR performance whether you think benchmarks are accurate or not. I never said the variance equated to a performance difference.

tuskenraider said:

I forgot I'm on the [T]ard forum where people make assumptions and put words into people's mouths. Simply put, I don't want a 10% drive variance in STR performance whether you think benchmarks are accurate or not. I never said the variance equated to a performance difference.

Click to expand...

But... but... If there's not actual performance difference, what does it matter what synthetic benchmarks show? If you bought a car to drive to the supermarket, and it performed poorly in the quarter mile, what relevance would that have to the supermarket task? I hope you can draw the analogy yourself.

I may just have to buy a pair of the 740adfd's and run tests myself. All anyone else will run is HDTach or similar synthetic benchmarks. I'd probably have to buy some games, too Such is the price of progress...

unhappy_mage said:

But... but... If there's not actual performance difference, what does it matter what synthetic benchmarks show? If you bought a car to drive to the supermarket, and it performed poorly in the quarter mile, what relevance would that have to the supermarket task? I hope you can draw the analogy yourself.

I may just have to buy a pair of the 740adfd's and run tests myself. All anyone else will run is HDTach or similar synthetic benchmarks. I'd probably have to buy some games, too Such is the price of progress...

Click to expand...

WTF?

This is getting dumb. HDTach is a great tool to compare one single drive or RAID set to another. I don't think anybody's trying to say that HDTach directly proves or disproves real world results.

Why is it that as soon as an HDTach picture comes up in the "Storage" section, people start turning into friggin' horse's asses?

And why the need to make personal attacks? - "Look who I'm talking to."

[T]ard Forum is right.

Brahmzy said:

This is getting dumb. HDTach is a great tool to compare one single drive or RAID set to another. I don't think anybody's trying to say that HDTach directly proves or disproves real world results.

Click to expand...

Except, say, when someone takes HDTach seriously enough that they RMA a drive because it doesn't do well enough in HDTach. The real world result there is, WD has to pay shipping to replace a drive, and the prices of drives go up ever so slightly.

Brahmzy said:

Why is it that as soon as an HDTach picture comes up in the "Storage" section, people start turning into friggin' horse's asses?

Click to expand...

Because people seem to think that HDTach means as much for hard drives as Sandra does for CPU and memory speeds, and that's simply not true. Hard drives are much slower, so they have a lot more time to strategize about how to go about accessing things. So any synthetic pattern simply doesn't show the real life performance of real life applications. With memory (for example) STR is important because pretty much any access pattern happens at that speed (yes, that's a sweeping generalization and I should be shot). With hard drives, STR is unimportant if you're not doing sequential transfers, and for most workloads, you're not.

Brahmzy said:

And why the need to make personal attacks? - "Look who I'm talking to."

[T]ard Forum is right.

Click to expand...

I apologize for making that remark. I'm pretty tired of this argument, too; I've been making my point for a long time, and been ignored for a long time.

But I think it kind of invalidates your point to "fight fire with fire". Yes, my attack was unnecessary; no, it doesn't help your case to call me retarded.

I tell you what, if you're truly interested in proving your case and not just arguing it, PM me and I'll send you links to some testing software I'd call a little more real-life based.

Except, say, when someone takes HDTach seriously enough that they RMA a drive because it doesn't do well enough in HDTach. The real world result there is, WD has to pay shipping to replace a drive, and the prices of drives go up ever so slightly.

Click to expand...

And a real world business has the cost of estimated replacement drives built into the current unit cost(Econ101). You want to bust my balls about returning a drive based on a couple of benchmarks fine, but I see it as the benchmark testing the theoretical performance of the drive. If doesn't even live up to spec, yet alone to another exact same drive on my system, why should I believe real world performance isn't going to suffer. And why should I waste my time further to prove so by measuring app loading times, etc. And again, the Raptor is a drive we page a HUGE premium on and so I don't think WD deserves as much leeway in performance tolerances. I wasn't on the phone being pushy, WD had no big issue with this, agreed with my reasoning and replaced it. If you demand less out of your purchases than what is advertised, well we have different standards.