Crucial DDR4 2133 MHz 32 GB (4x 8 GB)

[cadaveca]

Crucial is a name most of us know well. From memory to SSDs, Crucial has been a major player in the memory market for many years. Now that DDR4 is on the market, we turn to Crucial to get a look at what DDR4 has to offer.

Show full review

 

[KarymidoN]

I think I'm the only person in the world that will just migrate to DRR4 when launching DDR5? With these prices and minimal performance increase ... is not paying off move to DDR4.

 

[cadaveca]

I think I'm the only person in the world that will just migrate to DRR4 when launching DDR5? With these prices and minimal performance increase ... is not paying off move to DDR4.

DDR4 alone, sure. But there is not an option for DDR3 on X99. I have a bunch more kits to cover quickly in the next few weeks, so this is the start of looking into DDR4 performance, but I do feel that once you start to increase speeds, the change does pay off. It IS new tech though, and to me, is off to a good start. The transition from DDR2 to DDR3 was no different.

 

[BiggieShady]

I have a bunch more kits to cover quickly in the next few weeks, so this is the start of looking into DDR4 performance

Eagerly expecting to see what Samsung 20nm ICs can do
It's kinda weird to see you not even try to play with the timings, seems with DDR4 so early the best you can expect is for XMP profile to work correctly

 

[KarymidoN]

DDR4 alone, sure. But there is not an option for DDR3 on X99. I have a bunch more kits to cover quickly in the next few weeks, so this is the start of looking into DDR4 performance, but I do feel that once you start to increase speeds, the change does pay off. It IS new tech though, and to me, is off to a good start. The transition from DDR2 to DDR3 was no different.

I hope so, but I still think we will only have good DRR4 modules when we are close to the DDR5.

 

[IRQ Conflict]

The transition from DDR2 to DDR3 was no different.

What transition to DDR3?

 

[enikmaster]

In terms of temperatures, is there any difference between DDR3 and DDR4?

 

[Aquinus]

when we are close to the DDR5.

That's news to me. Care to elaborate on that and share some sources?

What transition to DDR3?

I was using DDR2-800 until 3 years ago, I mainly upgraded because I wanted more than 8GB of ram and the Phenom II 940 and DDR2 was grossly inadequate. There really aren't many tasks where I would want more than 4c/8t, to be honest and quad-channel DDR3 is more than enough bandwidth for a quad-core... or a 10 core Xeon. I could never be driven to upgrade for memory alone unless it's a capacity problem. I guess that's my real point.

On topic: Another great review, Dave. While I like DDR4, I still couldn't imagine investing in it until latencies get a little lower.

 

[Delta6326]

Great review Cad as always!

Also Screw DDR3 I'm going to DDR4 then next will be DDR6 and so on...

 

[cadaveca]

In terms of temperatures, is there any difference between DDR3 and DDR4?

That's too general. Some sticks offer thermal monitoring, but there are several different memory ICs available right now, and each is a bit different. So I'll have more on that particular subject in the near future.

 

[Folterknecht]

Cant find the OC results. Am I blind or ...

 

[RCoon]

Glad to see you back in business Dave

 

[Chaitanya]

This another company which has failed to live to its "legendary" reputation in my experience. Their Ballistix kits have left a bitter taste in my mouth. Apart from false advertisement of clock speeds on DDR3 kits, every single Ballistix kit I purchased in each generation of DDR (except for 4th gen) have failed to operate within reasonable expectation at stock settings.

 

[Axaion]

This another company which has failed to live to its "legendary" reputation in my experience. Their Ballistix kits have left a bitter taste in my mouth. Apart from false advertisement of clock speeds on DDR3 kits, every single Ballistix kit I purchased in each generation of DDR (except for 4th gen) have failed to operate within reasonable expectation at stock settings.

Really?, do you beat them with a hammer first?

 

[mroofie]

That's too general. Some sticks offer thermal monitoring, but there are several different memory ICs available right now, and each is a bit different. So I'll have more on that particular subject in the near future.

Those timings are quite ridiculous im sticking with DDR 3
What do you think dave ?

 

[cadaveca]

Those timings are quite ridiculous im sticking with DDR 3
What do you think dave ?

I think the memory isn't the problem, and the majority of people complaining wouldn't have bought it in the first place. The idea that timings are high, to me, is FUD. High compared to what? It's DDR4, not DDR3. Yet at the same time, we have the same CPU cores behind this new memory, too. So the idea that bandwidth at the moment is similar, when both technologies have similar products using them... well...uh...

Cant find the OC results. Am I blind or ...

OC results change from set to set. When it doesn't, I'll report about it. CPU and CPU cache are already OC'd. Best I got with this set was a bit over 2666 @ C13, 1.35V.

DDR4 OC is complicated. You cannot simply just change the first 4 timings and get results like I do. If I OC that way, the best these sticks would do was 2620 MHz C15. Other timings MUST be adjusted. So, do I show results with my tweaked settings, or with just adjusting the first 4, like most users will do, and then not be able to get the same results? How should I test stability of these OCs? Just enough to do the benchmarks, or enough to truly be stable? (hint: There's a difference)

OC results aren't something "based on fact". I report the facts (to the best of my abilities).

I WILL report a full rundown on memory OC in the future, separately.

 

[Chaitanya]

Really?, do you beat them with a hammer first?

Nah, had all sorts of problems with each generation of Crucial Ballistix I purchased.
DDR1 ballistix used to overheat and crash the PC up when Crucial specified timings and voltages were set. Also they burned after 4-5 months of purchasing those. Never got replacement for those kits to this date.
DDR2 Ballistix tracers developed errors causing BSODs after 3-4 months of use at stock settings.
DDR3 Had purchased the kit advertised to be 1600mhz, but it turned out be a 1333Mhz kit(I felt cheated at that point). Also when I manually set the timings, voltages and clock speed from BIOS, within 15 days of normal use one of the sticks failed causing BSODs and failed POSTs.

Apart from fooling customers with wrong specifications, I think Crucial support is a scam as its has been well over a month since I RMA'd the DDR3 Ballistix kit and I haven't heard back from those idiots to this date.

 

[Folterknecht]

I was just wondering about "Test System and Max OC" which I thought were missing. For me the headline was a little misleading it seems.

And I know that there is more to Mem-OC ... but thanks for your explanation.

 

[cadaveca]

I was just wondering about "Test System and Max OC" which I thought were missing. For me the headline was a little misleading it seems.

And I know that there is more to Mem-OC ... but thanks for your explanation.

Ha! +1 to you. That's a left-over from the past review template.

 

[Farmer Boe]

Another great RAM review Dave. I sure missed the OC section but understand the reasons for not putting that in at this time. Crucial is probably my favorite memory brand as I've never had one failure and every kit seems to overclock very well, no matter the platform. I look forward to your memory OC article!

 

[deemon]

What transition to DDR3?

Indeed. Still using DDR2-800 to this day also.

(and GTA V runs @ ~45fps @ high/very high settings @ 1080p ... on my 7 years old tech ... actually tought about upgrade like 4 years ago and decided that will do when intel comes out with next big thing.... still waiting.)

 

[deemon]

DDR4 OC is complicated. You cannot simply just change the first 4 timings and get results like I do. If I OC that way, the best these sticks would do was 2620 MHz C15. Other timings MUST be adjusted. So, do I show results with my tweaked settings, or with just adjusting the first 4, like most users will do, and then not be able to get the same results? How should I test stability of these OCs? Just enough to do the benchmarks, or enough to truly be stable? (hint: There's a difference)

Yes, please show results with all tweaked settings shown (not just the first 4). Also please post only truly stable results, as those are the only thing that really matter.