Wednesday, May 11th 2022

Sabrent Announces High-Performance, Low-Latency DDR5 Memory Modules

Sabrent Rocket 8 GB, 16 GB, and 32 GB DDR5-UDIMM's 4800 MHz Memory Modules for PC's. It seems like DDR4 has ruled the roost forever. Finally, new motherboards and architectures have allowed memory to grow wings with the advent of DDR5. Improved clock speeds promise up to twice the bandwidth of DDR4, all with lower voltage for better efficiency. To further help with that last bit, DDR5 has on-board power management to reduce motherboard cost and complexity. Sabrent's new DDR5 kits simply plug and play, install your memory kits of choice, enter the BIOS and hit AUTO, and reboot. For those that want to OC these kits, there is plenty of headroom to advance speeds if you want.

Reliability has also been improved by the introduction of on-die error-correction technology. Memory availability is improved by splitting the internal 64-bit (72-bit with ECC) channel into two independent, 32-bit channels (40-bit with ECC) and the addition of the same bank refresh. Combined with other features, including those that allow for up to four times the module capacity with consumer DRAM, DDR5 ensures your multi-core CPU won't be memory starved.
Our DDR5 comes in both DIMM and SO-DIMM form factors to help you out, whether you're using a desktop, laptop, or other embedded device. We meet JEDEC standards to ensure compatibility and reliability. We offer a wide range of capacities at an affordable price to make upgrading an easy choice - even the installation process is easy. So, what are you waiting for? It's time for the newest technology to grace your machine.

Sabrent Rocket 8 GB, 16 GB, and 32 GB DDR5-UDIMM's 4800 MHz Memory Modules highlights:
  • LESS VOLTAGE, BETTER EFFICIENCY: DDR5 requires less voltage, which results in lower power consumption than DDR4. On-module power management further increases efficiency and reduces motherboard complexity.
  • UP TO DOUBLE THE BANDWIDTH: DDR5 allows for up to double the bandwidth of DDR4, reducing a memory bottleneck. Content creation, gaming, and more will benefit from these rocket-fast speeds.
  • HIGHER PERFORMANCE: DDR5 has other internal improvements that lead to better overall performance. Make the most of your memory and reach the next level in gaming with these new technologies.
  • HIGHER CAPACITY: DDR5 can have up to four times the capacity in the same space, making sure you never run out of memory again. More is better, especially if you're juggling sixteen cores or more.
  • COMPATIBLE AND RELIABLE: DDR5 also has improvements to error correction, making sure your experience is consistent and reliable. Further, our memory meets all JEDEC memory standards, so it's easier than ever to upgrade and be on your way. Installation is always simple and easy, too.
Sabrent Rocket SB-DR5U-16G and SB-DR5U-32G details:
  • Lower voltage for lower power usage, cooler running
  • Better XMP for compatibility, better OC, dynamic = efficiency
  • Higher bandwidth, ideal for content creation
  • Performance is higher due to internal structure (gaming)
  • Higher density: 4x for client
  • On-die PMIC for better reliability, compatibility
  • ODECC for reliability
  • DDR5-4800/PC MHz/CL40
  • 280-pin UDIMM, 1.1V
  • 2x16/2x32GB Quad-Channel
  • Dimensions (LxWxH): 133.35*31.25*2.27mm
  • SB-DR5U-8G
  • SB-DR5U-16G
  • SB-DR5U-32G
  • Pre-Orders and pricing will be available shortly.
Add your own comment

47 Comments on Sabrent Announces High-Performance, Low-Latency DDR5 Memory Modules

#1
Daven
No latency numbers in the press release.
Posted on Reply
#2
bonehead123
Nice, clean-looking sticks,

but....

but....

Releasing RAM without any R.friggin.G.friggin.B.friggin circus effects....

HELLO Sabrent, this is illegal....hahahahaha :D




j/k of course !
Posted on Reply
#3
lexluthermiester
btarunrUP TO DOUBLE THE BANDWIDTH: DDR5 allows for up to double the bandwidth of DDR4
This has yet to be proven.
DavenNo latency numbers in the press release.
Right? Literally nothing stated..
Posted on Reply
#4
ir_cow
lexluthermiesterThis has yet to be proven.
AIDA64 says otherwise. Just depends on what DDR4 frequency is being compared DDR5
Posted on Reply
#5
InVasMani
Until I start seeing 64GB DIMM's that higher capacity marketing isn't rooted in reality.
Posted on Reply
#6
TheDeeGee
bonehead123Nice, clean-looking sticks,

but....

but....

Releasing RAM without any R.friggin.G.friggin.B.friggin circus effects....

HELLO Sabrent, this is illegal....hahahahaha :D




j/k of course !
D15 friendly <3
Posted on Reply
#7
lexluthermiester
ir_cowAIDA64 says otherwise. Just depends on what DDR4 frequency is being compared DDR5
Screenshots? Let's see some proof otherwise I call rubbish & bunkum. Every review I've seen shows DDR5 is not a big jump from DDR4. Even the review here on TPU shows a less than stellar(read almost non-existent) advantage and that was comparing DDR4-3200 to DDR5-6000. In a few testing cases some of the DDR4 handily beat the DDR5-6000 into the ground. So unless they've pulled a rabbit out of their hat, I call meadow-muffins on that double-the-bandwidth claim.
Posted on Reply
#8
SomeOne99h
DavenNo latency numbers in the press release.
lexluthermiesterRight? Literally nothing stated..
Scroll down.
  • DDR5-4800/PC MHz/CL40
Posted on Reply
#9
ir_cow
lexluthermiesterScreenshots? Let's see some proof otherwise I call rubbish & bunkum. Every review I've seen shows DDR5 is not a big jump from DDR4. Even the review here on TPU shows a less than stellar(read almost non-existent) advantage and that was comparing DDR4-3200 to DDR5-6000. In a few testing cases some of the DDR4 handily beat the DDR5-6000 into the ground. So unless they've pulled a rabbit out of their hat, I call meadow-muffins on that double-the-bandwidth claim.
Check out my DDR5 reviews on TPU. On the extremes is nearly double. 3200 vs 6800 for AIDA64 Read. If you go by the theoretical bandwidth that is a whole other topic lol.
Posted on Reply
#10
P4-630
I will be getting myself some corsair vengeance 36-36-36-76 @ 5600MHz...
Posted on Reply
#11
lexluthermiester
SomeOne99hScroll down.
  • DDR5-4800/PC MHz/CL40
That is not full spec data and is useless without the context of the rest of the timings. So really, nothing was said.
P4-630I will be getting myself some corsair vengeance 36-36-36-76 @ 5600MHz...
I'd prefer 32-32-32-68 as that seems to be a sweet-spot for DDR5, in theory.
ir_cowCheck out my DDR5 reviews on TPU. On the extremes is nearly double.
I provided a link to W1zzards review, where's yours? I'm not accepting the claim of "double" without supporting evidence.
Posted on Reply
#12
SomeOne99h
Not full spec yes, but isn't the first number is the most that makes difference in performance?
Posted on Reply
#13
ir_cow
lexluthermiesterI provided a link to W1zzards review, where's yours? I'm not accepting the claim of "double" without supporting evidence.
Why so lazy? Its easy to click on a review. Anyways here you go.

Posted on Reply
#14
lexluthermiester
ir_cowWhy so lazy? Its easy to click on a review. Anyways here you go.
It's not laziness, I'm a busy guy. Link not picture. I want to read the whole review. You're quoting a review but you're not linking it. You're showing one graph without any context that stands in direct contradiction to W1zzards review linked above, which carries credibility.
Posted on Reply
#15
ir_cow
lexluthermiesterIt's not laziness, I'm a busy guy. Link not picture. I want to read the whole review. You're quoting a review but you're not linking it. You're showing one graph without any context that stands in direct contradiction to W1zzards review linked above, which carries credibility.
Once again is it so hard to go to the memory section and click on any of the DDR5 reviews because you don't have time. But you do have time to reply and demand a link?
V-Color Manta XPrism RGB SCC DDR5-6200 CL36 2x 16 GB Review

Also what will reading the whole review tell you? "DDR5 provides double the bandwidth compared to DDR4". This of course must be in context. So to be clear, in AIDA64 comparing DDR4-3600 CL14 vs DDR5-6800 CL32 - it is nearly double. Go to DDR5-6800 and it is over that amount.
Posted on Reply
#16
lexluthermiester
ir_cowOnce again is it so hard to go to the memory section and click on any of the DDR5 reviews because you don't have time. But you do have time to reply and demand a link?
V-Color Manta XPrism RGB SCC DDR5-6200 CL36 2x 16 GB Review

Also what will reading the whole review tell you? "DDR5 provides double the bandwidth compared to DDR4". This of course must be in context. So to be clear, in AIDA64 comparing DDR4-3600 CL14 vs DDR5-6800 CL32 - it is nearly double. Go to DDR5-6800 and it is over that amount.
Yup, there it is, a cherry picked result. However, as expected, the claim of "double" is nonsense. I find it interesting that all of the other results are very much less than double. Seems to me like an outlier, especially given the massive number of other results and reviews out there. This is not the first time outlier results have been the cause of debate.
Posted on Reply
#17
ir_cow
lexluthermiesterYup, there it is, a cherry picked result. However, as expected, the claim of "double" is nonsense. I find it interesting that all of the other results are very much less than double. Seems to me like an outlier, especially given the massive number of other results and reviews out there. This is not the first time outlier results have been the cause of debate.
How is retail kits cherry-picked? I already said on the opposite ends of the spectrum. DDR4-3600 CL14 with 1T and Gear 1 is apex of DDR4. So is 6400 CL32 (as of now). If you go with less like DDR4-2133 vs DDR5-4800 is still a big gap. 32833 MB/s vs 76993 MB/s. Both JEDEC. Either way the claim that is is double bandwidth is value in both theoretical and actual real-world numbers. Does this translates to better performance for your system? No particularly, though that wasn't the argument you are making. This is a academic debate. Nothing more.
Posted on Reply
#18
lexluthermiester
ir_cowHow is retail kits cherry-picked?
The AIDA results are cherry picked.
ir_cowNo particularly, though that wasn't the argument you are making.
Correct. I'm taking issue with the with the claim of "double" performance, which is patently and plainly false, according to your own review. Is there an increase of performance? Sure. But double? Nonsense. Sabrent's claims are dubious at best. As are yours if you're going to cherry pick one single benchmark like AIDA.
Posted on Reply
#19
ir_cow
lexluthermiesterThe AIDA results are cherry picked.

Correct. I'm taking issue with the with the claim of "double" performance, which is patently and plainly false, according to your own review. Is there an increase of performance? Sure. But double? Nonsense. Sabrent's claims are dubious at best. As are yours if you're going to cherry pick one single benchmark like AIDA.
The press release doesn't state its double performance anywhere. Its 100% not true if that was the case. However saying double bandwidth is truthful and proven.
Posted on Reply
#20
lexluthermiester
ir_cowHowever saying double bandwidth is truthful and proven.
Clever marketing jargon. In the context of the article above, double performance is being implied.
ir_cowHowever saying double bandwidth is truthful and proven.
With one(1) utility. Care to use another utility that tests memory performance and update your review? Not everyone trusts AIDA as results rendered by that utility have been inconsistent at times over the years. This is why benchmarking is done with more than one testing method to begin with.
Posted on Reply
#21
ir_cow
lexluthermiesterClever marketing jargon. In the context of the article above, double performance is being implied.
You are implying double performance. Thats on you. Once again press release says nothing of the sorts.
Posted on Reply
#22
P4-630
lexluthermiesterI'd prefer 32-32-32-68 as that seems to be a sweet-spot for DDR5, in theory.
Well 2x16 GB 36-36-36-76 DDR5 5600MHz seems to be about in the middle of the graph @ir_cow posted...
I consider 32GB DDR5 5600 cl 36 @ 300 Euros good enough.....

Edit: Just checked what your "sweet spot cl-32" memory costs in my country....
There's only 4 sets of 2x16GB I can buy in my country:



o_O
Posted on Reply
#23
ThrashZone
Hi,
Sorry but 4800c40 doesn't sound very low in the latency department to me, seems very high unless it's 8000 :kookoo:
Posted on Reply
#24
lexluthermiester
ir_cowYou are implying double performance. Thats on you. Once again press release says nothing of the sorts.
Nice side step there. You trying to say that the implication I've suggested was not the intention of the marketing effort? IF you really think that, I've got a bridge in Brooklyn, NY I'd like to sell you...

However, try answering this;
lexluthermiesterCare to use another utility that tests memory performance and update your review? Not everyone trusts AIDA as results rendered by that utility have been inconsistent at times over the years. This is why benchmarking is done with more than one testing method to begin with.
Hmm...?
P4-630Edit: Just checked what your "sweet spot cl-32" memory costs in my country....
There's only 4 sets of 2x16GB I can buy in my country:
Keep in mind, we're still in the early stages of DDR5 lifecycle, performance will improve, as will timings and the prices will come down.
Posted on Reply
#25
ir_cow
lexluthermiesterCare to use another utility that tests memory performance and update your review? Not everyone trusts AIDA as results rendered by that utility have been inconsistent at times over the years. This is why benchmarking is done with more than one testing method to begin with.
Hmm....?
I'm going to make this simple to understand. I am not talking about performance of DDR5.

I simply pointed out the statement that DDR5 is double the bandwidth is correct. It seems you disagree and committed to saying its isnt.

Go back to read the comments if your still confused about this conversation. I gave you "proof". Nothing more to discuss here.
Posted on Reply
Add your own comment
Nov 21st, 2024 10:36 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts