Any reason to recommend the G.Skill Phoenix Pro?

A lot of people recommend intel,OCZ,Corsair,Samsung but I don't think I've seen anyone recommend G.skill, is it not a good drive? According to reviews it's as good or even better than the Corsair Force..Also the reliability, how is it? Cause here were I'm from, the Phoenix Pro is the cheapest option between Intel x25-m and Corsair force..I could probably get my hands on other brands but I don't think it's worth the trouble since these 3 shops offer shop warranty which will save me the hassle of having to have the drive delivered to another country in case rma is required..Not that big of a price difference though..around $10 to the intel, and another $10 to the Corsair Force..all 120gb

 

I've also find intresting to hear feedback on this. I've able to find this drive in Norway, and actually asked the almost same question in one of my other therads but got no direct answer...

 

Really? I didn't see that thread..Well hopefully we can get a more direct answer here.. It's the cheapest drive here plus the 1 yr shop warranty makes stuff a lot easier..other than sending it back to another country, the store will handle the rma process for you..So I am looking forward to real world reviews on this..It's just weird that I've never heard anyone recommend it..

 

Same here. However, I read that the new intel drive is on its way to Norway, I might go for that instad of x25m.

 

G.Skill Phoenix Pro is another drive based on the Sandforce SF-1200 controller. It will perform about the same as any other Sandforce SF-1200 drive... OCZ Vertex 2, G.Skill Phoenix Pro, Corsair F-series (Force), Mushkin Callisto, Patriot Inferno, etc. They all perform equivalently in real-world scenarios.

The reason that you haven't heard of the G.Skill Phoenix Pro is because it has smaller distribution than a company like OCZ, Intel, or Samsung. It doesn't mean it's a bad product... they just didn't spend as much on marketing than other companies.

If you can find a good deal on a G.Skill Phoenix, then get it. Because honestly, you will never be able to tell the difference between that drive, or any other Sandforce SF-1200 drive.

 

Would you recommend Sandforce drives over the intel when it comes to gaming and loading apps?

 

For your situation, they are about the same.

If you can get a better deal on a Sandforce drive, get that. If you can get a better deal on an Intel drive, get that. They are all good drives, perform equivalently well, and have been used by computer enthusiasts since they came out.

So get whichever one you can get a better deal on.

 

G.Skill only sells to online retailers I believe, so their market share is obviously going to be extremely limited compared to Corsair/OCZ.

As it's been stated, anything with that SandForce badge is now seen as a budget SSD while still retaining high performance (I own 2 SF SSDs).

 

http://forum.notebookreview.com/solid-state-drives-ssds-flash-storage/564939-comparison-chart-ssds.html

 

How long have you been using your sandforce ssds?

 

My Agility 2 60 GB has been going strong in my Vostro 1500 for 9+ months, no hiccups/freezing to report.

My Vertex 2 80 GB has been in my Latitude 13 for 2+ months, again no issues.

 

Okay..I might go for the g.skill then, if it's just the same as the force series and the force is around $20 more expensive..But then theres the intel x25-m which is virtually the same price lol..This is hard..

 

My advice to any potential SF buyer is still, check to see if there are people using your notebook model with a SF drive(brand doesn't really matter) for say 3 months+ to rule out compatability issue.

Getting a lemon at the start of the gate is not a big deal assuming you have good return policy(this thing happens). Getting a drive which is not compatible but only being shown after a short period of use(thus needing to go through the restore process and the interruption) would negate any milisecond gain(or price gain of ~20 bucks max), IMO.

 

That's not good..I have never heard of anyone use a G.skill Phoenix Pro..much less someone with my exact system..Maybe I should just go for the x25-m and save myself the trouble of "gambling" whether a drive will work or fail..

 

First thing to check is OCZ Vertex 2/Agility 2. If there are people using them in M11X R1 and are happy with it, you would most likely as well. The other is Corsair. Just go over to their forums and ask if there are happy or unhappy users and you can get a clue.

My undestanding is that SF drive have a lot if issues with older Dell business lineup(which OCZ didn't admit initially because they didn't even bother to test but now do and as usual blame the BIOS or whatever).

 

Is the performance of the intel far from the sandforce drives? i mean with regard to my usage which I believe doesn't require much sequential write performance..Cause I believe the shift of OCZ to 25nm isn't going so well performance wise,and I don't wanna join in the fuss..

 

Performance wise, I continue to say they(all last gen SSD) are more or less the same(thus millisecond gain/loss) unless you have a very specific usage pattern which you know would be benefited from a specific model.

Does loading Excel in 5s vs 5.5s really means that much(that would be further blur by Superfetch etc.) ?

that is the reason why I keep X25M G2 way above the list in that smooth workflow(including that there is very little chance you hit a bad drive of x25m) trumps millisecond gain(if there is any) here and there.

 

The Intel X25-M is a fine choice. You'd do well with it.

But to be quite honest, I think Chimpanzee's post was a little bit of a "boogeyman" post. Yes, you should ensure that the owner community for your laptop model doesn't have known problems with an SSD model. But there are PLENTY of people running Sandforce SF-1200 SSDs on Alienware M11x R1s and R2s. It is a non-issue for you.

(1) No, performance between Intel X25-M and Sandforce SF-1200 isn't far off in real-world performance for most users. You will have to be in very specific situations, or running synthetic benchmarks, in order to be able to tell any difference whatsoever.

(2) The whole OCZ fiasco is based on OCZ not properly communicating the nuances from using NAND flash memory manufactured on a 25nm transistor process, rather than a 34nm transistor process.

Usable capacities dropped. Performance dipped a bit. But these aren't anything really worth getting upset about. What made people upset was that they bought OCZ Vertex 2 drives, and received products that had these changes, without anyone ever communicating this to them. Every other SSD manufacturer must go through the same thing when they go from 34nm --> 25nm, and they all learned from the OCZ fiasco to clearly communicate these changes.

So, the OCZ fiasco is a problem dealing with communication of these changes to the customer, and the resulting response where OCZ essentially acted like jerks... The fiasco didn't get the atttention that it did because of the product(s) themselves.

Since then, OCZ has been the whipping boy for all kinds of internet rage. Ask the right angry nerd, and they'll tell you that OCZ started the Chicago Fire of 1871. Having said that, I still wouldn't recommend you buy OCZ products. Even after this whole fiasco blew up into a public relations nightmare, they were jerks about "making it right" with their upset customers. That is not the kind of support that you want as a customer.

An SSD is not about application load times or boot times. Any drive, including mechanical HDDs, can load applications relatively quickly. It is about multitasking performance.

When you're throwing multiple read/write requests at an SSD, it won't choke like a mechanical HDD will. You will never be sitting there with a sluggish computer, listening to the mechanical HDD thrash away as it tries to keep up. Everything you do will just seem to "pop" and "snap" almost instantly.

The closest analogy I can think of is internet connections. You're probably using a fast, broadband connection right now. Web pages "pop" up. Things you do on the internet are smooth and fast. Now, imagine if someone replaced your fast broadband connection with a 56K dial-up modem, and how that would make everything you do on the internet feel sluggish and slow. That is the kind of difference you can expect from SSD vs mechanical HDD.

But you can't look at an SSD, and try to justify it in terms of time-money saved, because you save X amount of time waiting, and your time is worth $Y/hr. The reality is that SSDs are luxury performance items to most people that can afford them. Very few people NEED a Ferrari. But it sure is fun to drive one.

 

Also, the G.Skill drive is probably using 34 nm NAND, and not 25 nm (or if it is, they should clearly state so). This means that it shouldn't be suffering the performance drop-offs that the switch to 25 nm NAND in the unmarked Vertex 2s did (which is the other major reason for the blowup against OCZ).

 

What do you mean 'boogeyman' post ? Did I say it is going to cause him trouble ? Didn't I say it just check with other users with the same model ?

 

But if I had a choice between intel and sandforce..I would seriously not feel any difference between them? Like I could just flip a coin? Well i guess the g.skill retailer here has a shop warranty..but then it would be great knowing that my intel ssd won't fail on me no matter what I do to it..

 

Check the review from Laptopmag. It shows you exactly how much difference there is between Sandforce (Vertex 2) and Intel X25-m.
Intel SSD 320 (300GB) Review - A Review of the Intel SSD 320 (300GB)

 

Oh my..The real world numbers are really close..I mean you weren't kidding about not being able to tell the differences between them..They (at most) 1 second apart..maybe 2 but still..So I guess I'm going for intel..

On a side note, the new series of intel 320s aren't really "that" big of an improvement from the x25m..I mean the only noticable difference is probably a faster copy paste speed..I mean the sequential writes are a good improvement but not really an actual "improvement"..Will an x25-m user notice the difference if he/she gets a 320? i mean it is faster but I'm guessing that has more to do with the drive size of 300gb rather than the actual performance..and even that difference isn't that big..comparing a 120gb to a 300gb I mean..

Another thing to note would be the Vertex 2 numbers are most probably the 25nm drive numbers..Now I don't really know how big the difference between the 34nm and 25nm Vertex 2 drives are, but I believe the 25nm performs less than the original 34nm ones..So maybe the G.skill Phoenix 2 Pro with the 34nm sandforce drives will actually perform better than the Vertex 2 on the benchmark shown above..Hmm, it's all very confusing lol..

 

(1) There will not be a real-world difference in performance between an Intel X25-M and an Intel 320.

(2) Where you WILL find a real-world difference is that you can actually buy an Intel X25-M SSD. Intel 320 SSD's are going to be hard to find and command a price premium for the near future.

 

I agree with you 100%..The best "feature" of the 320 is that it'll hopefully lower the prices of the G2.. But I hate how some people would say the 320 is "way better" than the x25-m just because it's newer and intel says it's better lol

 

And you won't notice a difference in real life between a 34nm Vertex 2 or a 25nm.

 

Really? but the numbers benchmarks on the 25nm Vertex 2 and the 34nm vertex 2 are halved..I would think that'd make a big difference in real world applications?

 

In certain synthetic benchmarks the performance was halved.

There's a big difference between real world performance and synthetic benchmarks.

 

It depends on the particular benchmark. The benchmark that really matters in real-world performance for desktop / laptop users is Random 4KB Reads at low queue depths. The OCZ Vertex 2 25nm is slower than a 34nm drive, but not enough to make a difference.

 

As far as I know the Kingston V+ scores quite bad at 4K random reads, yet it's real world performance is very good.

In my opinion 4K random reads aren't a very good predictor anymore for real world performance.

 

I'm not disagreeing with you, but could you explain a bit more about why you think that?

 

It seems to no longer matter above a certain treshold. Anand said something similar about the Intel 510 I believe. 4K random performance was not so good, but it didn't have negative impact on real world performance.

The Kingston V+ is the most striking example. If you check out the 4K random reads it looks like a bad drive, yet it's real world performance it beats many other SSDs (Techreport.com).

 

Ah. I think you may have misread that part. In the Conclusion (Final Words) of Anandtech's review of the Intel 510 SSD;
The Intel SSD 510 Review - AnandTech :: Your Source for Hardware Analysis and News

" Random read performance, as it turns out, has a pretty major impact in the real world."

Random write performance is also pretty low by today's standards, however the impact on most of our real world performance tests is minimal. It looks like we may have hit the upper limit of what we need from 4KB random write performance (at least given current workloads).​

 

What I read was a combination of Anand and Hardwareheaven on this page:
http://www.hardwareheaven.com/reviews/1130/pg13/intel-510-series-250gb-ssd-review-conclusion.html

So the Vertex 3 has double the 4K random reads of the Intel 510.

Now compare the real world performance of the Vertex 3 and Intel 510 here:
Crucial m4 256GB SSD (C400) Review - File Copy Tests (from there on 5 pages)

Very small differences. I believe if you add all the seconds up it works out as something like 1% difference on 800 seconds. So a 100% difference in 4K random read performance and a 1% difference in real world performance.

That's why I say 4K random read isn't a very good predictor for normal usage. 4K random reads are more interesting for extreme multi tasking or multiple user environments.

I don't know if you've looked into Kingston v+, it's a good example too. Relatively poor 4K random performance, good real world performance.

 

If I remember correctly: 870 vs. 874 (Intel vs. V)... so less than ~0.46% difference actually.

Yeah, there are other things than single number 'scores' that determine the performance of any drive (SSD or HDD) in actual / normal use.

 

I don't know if you saw this, but Anandtech published a correction. The ridiculously high numbers for the OCZ Vertex 3 were an error:
Correction: OCZ Vertex 3 Random Read Performance Data - AnandTech :: Your Source for Hardware Analysis and News

I would not consider a file copy to be a good example of real-world performance, because nobody buys an SSD to copy files from point A to point B faster. They buy an SSD to run applications faster, especially when multitasking, which hits the SSD with multiple I/O requests. And in that type of scenario, Random 4KB Read speeds (at multiple NCQ depths) are the closest benchmark measurement tied to performance in that kind of usage pattern.

A lot of review sites are now showing Random 4KB Read speeds for both low and high NCQ depths, which is a good indicator to show how drive performance scales as you pile on the data requests in multitasking scenarios.

Don't get me wrong - I absolutely agree with your view that Random 4KB Reads aren't the be-all-end-all benchmark number to indicate real-world performance. In reality, once you get an SSD that has "good enough" performance, you quickly start hitting dimishing returns for your money. You could buy an SSD that costs 2x as much and has 2x better Random 4KB Read performance, but yields only marginal real-world benefit. We both agree here.

The only reason I am bringing Random 4KB Read speeds up is because a lot of people who look at HDD vs SSD only pay attention to Sequential Read speeds, because that's the "big" number. And they might think that an SSD is marginally better than a HDD, because Sequential Read speed is "only" 2x - 3x faster (80-100MBps --> 250MBps). But in reality, it is the Random 4KB Read Speed indicator that really shows the benefit of SSD vs HDD, where the speeds are 100x - 150x faster on an SSD (0.5MBps --> 50+MBps)

 

I did saw that yeah, but as I have very little interest in synthetic numbers it didn't really register in me.

So the 4K random read is still 80% higher than the Intel 510.

It's a small part of real world performance. That's why I said including the 5 next pages with all kind of real world performance examples.

I get what you're saying and I mostly agree. I used to be more interested in 4K random performance untill I noticed that the Kingston V+ booted my system faster than any other SSD I tried, including the Vertex 2 with it's fantastic 4K random performance. Now I'm only interested in real world performance measurements.

 

The benchmarks also indicate the 4K random scores for the 25nm Vertex 2 is half of what the 34nm Vertex 2 used to be..Okay so if even the 4K doesn't impact real world performance 100% accurately, how are we supposed to gauge the "performance" of these ssds then? I understand the whole sequential marketing thing and all, but I always thought I could trust the 4K randoms to provide an accurate measure of real world performance..

 

The only way to gauge accurately is real world benchmarks.

 

Wherein ssds have, at most, a 2 second difference in loading times with each other..(SATA 3 not included) So it really does fall down to pure preference and brand loyalty..As well as reliability and price..Which makes choosing an ssd that much more difficult..

 

You're probably looking at a very specific type of benchmark scenario (Random 4KB Read Speeds at a particular NCQ depth, or QD).

Random 4KB Read Speeds matters when trying to show the difference between HDD --> SSD. People who looked only at the "big" Sequential Read Speed numbers between HDD --> SSD didn't get the whole story (only 2x-3x speed increase). They needed to look at the Random 4KB Read Speeds to really understand what was going on (100x-150x speed increase).

When you're comparing modern SSDs, you are already beyond the point of diminishing returns. Random Read speed differences are going to be well below one order of magnitude between modern SSDs... nowhere near the 2x orders of magnitude difference between HDD --> SSD.

Phil is correct - the only way to really tell real-world performance is real-world benchmarking. But even there, you won't really see much of a difference. Once you get an SSD that is "good enough", you really won't see much difference in real-world performance. And any modern SSD, including the ones mentioned in this thread, will all be "good enough" for all but the most demanding usage patterns.

 

The two major factors to consider are price and capacity. The price-per-GB ratio is the biggest differentiator when you're looking at these "good enough" SSDs.

 

The thing is though..Here where I'm from, the price of 120GB intel, 120GB G.skill, and 120GB Corsair are about the same..around $10 difference from each other, the corsair force being the most expensive and the intel being in the middle..I'm considering either the G.skill Phoenix Pro since it's the cheapest or the intel x25-m since it's proven to be reliable..So the reason I'm having so much difficulty deciding is because there are virtually no differences between the drives..

 

Flip a coin?

Seriously, you'll be fine with any of those drives.

 

Was considering that too! Maybe I will..Heads for intel tails for G.skill lol..But maybe going with intel due to reliability..=)