Date: September 22, 2006
Author(s): Rob Williams
With so much competition in the memory market nowadays, it’s increasingly difficult to make the -best- choice when it comes to purchasing a new kit. The Crucial PC2-6400 use the infamous D9GMH chips, but requires a staggering 2.2V! Does its performance make up for it?
It has been quite some time since we’ve taken a look at a Ballistix kit. Over a year, in fact! The Ballistix PC3200 kit was actually the first ram kit review we had, so looking at this one is like deja vu. Back then, that kit had offered great performance and good overclocking ability. I am hoping to see a repeat of those results today.
Despite the fact that Crucial is a huge stature in the US memory market, they are not that much of one in the enthusiast circles. Don’t get me wrong, they have their fans like any other company, but most people will first consider running to the likes of OCZ, Corsair, G. Skill and others. The primary reason I believe this to be the case is that Crucial doesn’t advertise their gaming kits like those other companies do. More times than not when I see a Crucial ad, they are advertising SD cards instead. Quite hard to overclock those.
Crucial is still very much important to performance memory though, but its due to the company behind them. Micron is the company responsible for all of the chips in the Crucial line-up, but also the majority of enthusiast modules on the market. Out of the last five memory kits I have taken a look at, four of them were using Micron D9 chips. The question remains… does Crucial save the best chips for themselves? We may just find that out today.
Crucial packages their modules unlike any others I can think of. While most stick to the popular plastic blister pack design, Crucial packages each individual module in an anti-static package, and then sets them inside a small box. Inside the box is a small divider, so that the modules don’t go flying all about. While this is a different idea, it’s designed to keep the modules secure until you get your hands on them.
Inside you will find the modules, an installation guide in addition to a coupon for iolo System Mechanic 6. The offer is to get the program for only $10, or $30 off the regular price. We don’t recommend anyone to take advantage of this offer unless you are willing to put up with a borked machine. Both Matt, myself and others on our forums have run into problems with the product.
The modules themselves use the classic Gold Ballistix heat spreaders. The only other Crucial modules that use different spreaders are the Tracer series.
I think the main reason Ballistix modules stand out in the crowd is the fact that they use a black PCB, as opposed to a green one which most others use. This adds a lot to the style in my opinion. Here are some additional product shots:
Now that we have you introduced, we can head on into the technical information and testing methodology.
The chips used in these modules are none other than the Micron D9GMH. Hmm… Micron chips in Crucial modules.. so odd! </sarcasm>
These modules are rated for DDR-800 speeds with 4-4-4-12 (CAS-tRCD-tRP-tRAS) timings, at 2.2v. 2.2v?? Yes, 2.2v. While other comparable kits on the market could suffice with 1.9v or even 1.8v at those timings, these are stickered for 2.2v. However, it was my main goal to see if that could be tweaked down a little bit.
The modules have the stats we are looking for, but the overclock ability is the primary concern here. The fact that stock ‘requires’ 2.2v doesn’t exactly give a reassurance that they will be that overclockable, but there is only one way to find out!
So, after a 3:00am run was over, I came up with these conclusions:
I spent about four straight hours trying to get the memory to be 100% stable at stock speeds with lesser than 2.2v, but it was not happening. At 1.9v, things were smooth, but not for long. It benchmarked okay, but Super Pi would crash immediately. At 2.0v, 3D Mark 01 would halt every few seconds.. sometimes not snapping back out of it. 2.1v returned the Super Pi problem. So, 2.2v on the testing motherboard proved to be a requirement.
450MHz was just as stable with 2.2v though, which I found surprising. I had to boost it up to 2.3v at 500MHz speeds which is not that much of a surprise… as it’s normal of other kits. Finally, 550MHz with 5-5-5 timings were completely stable with 2.5v… rather modest. So, I am impressed with the overclocking potential of these, because I didn’t expect it due to the fact that they required 2.2v for stock. Trying 4-4-4 timings at 550MHz was not happening, however. I tried it up all the way to 2.8vdimm, and it was still not stable.
Here is where I start to be confused though. Our friends at EclipseOC had a similar kit tested a few months ago, and they had far better results than us. So, I could have received a less than ideal kit. This could be considered a good thing, really. It goes to show that your mileage may vary, and it’s evidenced here.
With that out of the way… results!
Throughout all of our benchmarks regardless of what we are reviewing, testing is done in a clean and stand-alone version of Windows XP Professional with SP2. Prior to testing, these conditions are met:
If you are interested in using the same benchmarks as us, feel free to visit the developers website:
The testing rig used for today’s benchmarking is as follows:
I am pitting the Ballistix up against the recently reviewed OCZ 7200 EPP modules, which are based on the same Micron chips.
First up is Sandra and we see good results all around. However, the 7200 modules beat about the Ballistix in each and every round.
The same results are here… Ballistix falls just a little behind.
Everest proves a little more interesting. Though the first two sets of results declare a clear winner, the latter two are far closer to one another. That’s what I like to see from two sets of modules at the same speeds that use similar chips :-)
The latency tests give us even more varied results than the Read/Write ones. While latency doesn’t matter -that- much in most applications, the top overclock proved 1.5ns apart in the 7200’s favor.
Like in the Everest Read/Write tests, we are seeing something similar happening. The Sciencemark results for example, prove the 7200 to be the faster kit in the first few settings. However, 500MHz and beyond, the results are much tighter between each kit. The PC Mark results are the most interesting by far though. At the first two settings, there is a huge differential between the two… this time in the Ballistix favor. I can’t begin to understand why the huge difference, but it’s there. The final two settings proved close, but the Ballistix took the cake.
The same goes for the 3D Mark tests.. Ballistix won!
Let’s move onto our Super Pi results, and also our conclusion.
Finally, we have some Super Pi action. Long story short, the OCZ modules proved faster, but not by a huge margin. That’s to be expected in all Super Pi tests, really.
When I first laid my grubby hands on these modules, I had expected quite a bit. For one, they are Crucial! Two, they have D9GMH chips. I naturally expected a lot of performance and overclocking potential. While I am not incredibly impressed here, these modules do deliver what they promise.
Comparing these modules to others on the market in the same price range, it’s easy to find one better. For instance, the G. Skill PC2-6400 PHU2-2GBHZ have been making its rounds and pleasing many along the way. They retail for around $260US and have a ton of tweaking potential… much moreso than what I have seen with these Ballistix. I perused around a few different e-tailer sites, and one was charging around $330US for this kit, while another was charging $270. You will want to shop around, because I am sure you could even find them cheaper.
So, what would I recommend? If you are a Crucial fan and are partial to their modules, then this is a decent kit. However, 2.2v being the stock voltage is a little hardcore. If you are planning to overclock, then it won’t be a problem. DDR2-900 was just as possible without increasing the voltage… and at the same timings.
If you are not partial to Crucial, this I would avoid this kit. The performance and overclocking potential this kit offers just doesn’t warrant a $270US price tag. $240 would be more reasonable… but of course its the e-tailer who ultimately decides how much to charge. Even at $270, they will sell. Personally, I am a huge fan of OCZ’s kit that I have been including in my reviews lately, but they retail for $330-ish. If I were to run out and purchase a kit around the $250 area though, I would be quicker to jump on the earlier mentioned G. Skills over anything else.
Better chips would have made all the difference in these modules I feel. The stock voltage of 2.2v wasn’t just set by Crucial, it is actually mandatory to retain 100% stability. That’s rather high if you are planning to do no overclocking at all. This is actually the first kit I’ve come across that hasn’t worked on less voltage than noted… unless we count value sticks. But, that’s not what these are :-)
When it’s all said and done, I am giving the Ballistix PC2-6400 a 7 out of 10. Crucial… stop giving away all the best chips to other companies and slap them in your own modules! We beg you ;-)
If you have a comment you wish to make on this review, feel free to head on into our forums! There is no need to register in order to reply to such threads.
Copyright © 2005-2019 Techgage Networks Inc. - All Rights Reserved.