Introduction, Specifications, and Pricing
G.Skill is a company that needs no introduction, as it is more than likely that at some point in your computing that you have owned at least one set of their memory. Whether it be the various kits of Trident, Trident Z, Ripjaws with different Roman numerals after them, Flares, Snipers, Aegis, Ares, Pi, well, you get the point. If you are looking for RAM, G.Skill has enough options to sell you to fit the need. Even before we ventured into this review gig, we were well accustomed to what G.Skill offered and have bought many a set for multiple systems.
Even though the DDR4 market has moved towards looks over all other things, where fancy heat spreaders seem to sell RAM these days, who could forget the RGB craze that is sweeping the market! There are still users who prefer that their memory does its job without being the standout bit of kit in the chassis. Believe it or not, manufacturers do make DDR4 as they used to, where performance and capacity are the first things companies consider. The heat spreaders on the kit are likely more for series recognition than to remove heat or block EMI, which is why heat spreaders came into fashion in the first place.
One of those kits is what we have in hand, which just so happens to be one of the latest sets from the Ripjaws V series from G.Skill. We have reviewed and used many of these kits in the past, but we can say for sure that we have never used or tested a 64GB kit of RAM before this! While many will scoff at the density of this kit of memory, there are many out there with the need for more RAM when it comes to productivity and not gaming. Many a gamer will buy all the density they can afford and are not above opting for 32GB or 64GB sets for a gaming rig, but most times, that density is spread over four or eight sticks. With what we have in hand at the moment, we have 64GB of density in just two sticks, which is not only cool but leaves room on most motherboards to add even more density, should the need arise!
We borrowed the above chart from the product page at G.Skill, and in it is all of the information one might need to make an educated decision as to if it is the right kit of DDR4 for you. At the top, we are shown that this is a kit of DDR4 with 64GB of density in two sticks in this dual-channel kit. Using the built-in XMP profile, these Ripjaws V will run at 3200 MHz with 16-18-18-38 timings, drawing 1.35V. Interestingly, the SPD profile has this kit botting up at 2666 MHz with 1.20V of power needed. We also see a mention that no fan is included, these kits are backed with a limited lifetime warranty, and stabili8ty of the XMP 2.0 profile can be dependent on the CPU and motherboard used.
Also, on the same product page, we see the description of them, where G.Skill says this. "As the latest addition to the classic Ripjaws family, Ripjaws V series is the newest DDR4 memory designed for maximum compatibility and cutting-edge performance with the latest Intel Core processors. In this review, we plan to test its compatibility, not just with Intel, but with AMD as well, and we also test performance, and we will soon see how cutting-edge this performance is!
However, if the past has taught us anything, more density typically means slightly slower RAM with a tad more latency involved, as it is much easier to run 16GB worth of ICs than it is to run 64GB. Even so, we have high hopes that these new Ripjaws V kits from G.Skill can compete head to head while offering four times the density of what the average user will buy!
You may notice that in our provided links to buy this RAM, we had to show the 32GB variant of these Ripjaws V because Amazon does not have a listing of the exact set we are testing, not even with red heat spreaders, versus the black ones we were shipped. However, we do not stop there; we scour the internet to find them for a real-time price reference. We then went to Newegg, where we saw our exact kit listed, and to be real honest; we do not feel that the $229.99 pricing is out of line.
For a 64GB 3200MHz kit, you will be spending roughly $75, and with these Ripjaws V, we get four times the density, which when the math is done, comes to $300! Not only do you save around $80, but we also are not populating more than two slots on a motherboard with the Ripjaws V 3200MHz 64GB DDR4 we are about to show you!
Packaging and G.Skill Ripjaws V
While the packaging for these Ripjaws V sticks is the same as we have seen before, the cardboard insert inside the clear plastic packaging attracts attention with the bright red company name and the blue/white in the middle. Resting on the cardboard and snapped into the plastic are our two sticks of the all-black DDR$, and we can even see the case badge down by the Intel compatibility icon.
Flipping the packaging onto its face, below the Ripjaws V naming at the top, G.Skill explains how these kits ate designed for performance and stability, are hand-screened, ensuring reliability. Next, we see mention of the warranty, with a gap until we see the G.Skill information section next to the product sticker. We see the F4-3200C16D64GVK model number along with the speed, timings, and required voltage on the sticker.
With the kit out of the box, we can get an unimpeded look at them with their black PCBs and finely lined metal heat spreaders. The heat spreaders have wings on either end, where the middle takes on a V-shape, which goes well with the Ripjaws V series name. The series name and manufacturer are found on a sticker that is placed in the center. To this kit's right is a case badge, which can now be seen in its entirety.
We found the sticker on the other side of the sticks; when flipped around, we find another sticker. This time this sticker delivers the part number of the kit and a mention of voiding the warranty if you remove these stickers. We also see the date of manufacturing, speed, density, timings, voltage, XMP 2.0 readiness, and even each stick's serial number.
When looking at the top edge of the Ripjaws V, we can see the thin gap where the two halves of the heat spreader come together, as well as the tabs that lock them together. Along with the various heights of the sections, the angles used, and wider gaps between the sections, we see that the G.Skill name has been painted onto the center edge.
Opening Thaiphoon Burner to see what ICs are under the hood, we see 16 ICs under the spreaders, with eight of them on either side of the 8-layer PCB. Hynix makes the ICs, and while there are "?" in the model, we stripped the spreaders to see that these are H5ANAG8NMJR-VKC and are 2666 MHX downbin chips.
Installed on our AMD based system, we cannot get over the density we have to use with just two sticks. On this system, we still have room to add another kit for a whopping 128 GB of DDR4, well if the board and CPU can handle that amount! With a couple of white pops and a hint of red, these black modules blend in with the build.
Now in the Intel system, we have them installed next to each other without skipping a DIMM slot. Since the APEX offers two, we had no choice, but it does show what the Ripjaws V will look like when populating all slots on a four DIMM motherboard.
Test System Details
To obtain the AMD CPU-Z screenshots, you will see directly following this image, and this is the system we used to do it, as well as in attaining the results seen in the following pages. Thanks go out to Corsair, ASUS, and GIGABYTE for supporting this venture. For detailed specifications of the system, those can be found below.
After installing this DDR4, we spammed the delete key to enter the UEFI to enable the DOCP profile. Once completed and botted, the Ripjaws V come to life as we expected. We have them at 3200 MHz with 16-18-18-38 1T timings with 1.35V auto-set to the VDIMM and 1.088V for the SOC.
We attempted to reduce the timings while adding voltage to the tune of 1.45 VDIMM and 1.18125V for SOC. After a bit of trial and error, we ended up with 3200 MHz with 14-17-17-38 1T timings, which is impressive considering this kit's density!
We then attempt to push the kit's speed, using the same voltages described in the reduced timings run. While we did not get very far, we are impressed this amount of density allowed us to run these Ripjaws V sticks at 3333 MHz with DOCP profile timings.
Chad's AMD DDR4 Dual-Channel Test System Specifications
- Motherboard: ASUS Crosshair VIII HERO Wi-Fi - Buy from Amazon
- CPU: AMD Ryzen 9 3900X - Buy from Amazon
- Cooler: Corsair H150i PRO - Buy from Amazon
- Video Card: GIGABYTE GeForce RTX 2060 SUPER Gaming OC 8GB - Buy from Amazon
- Storage: Corsair Force MP500 480GB NVMe - Buy from Amazon
- Case: Thermaltake Core P5 TG - Buy from Amazon
- Power Supply: Corsair RM750x 750-watt - Buy from Amazon
- OS: Microsoft Windows 10 Home 64-bit - Buy from Amazon
To obtain the Intel CPU-Z screenshots, you will see directly following this image, and this is the system we used to do it, as well as in attaining the results seen in the following pages. Thanks go out to Corsair, ASUS, and GIGABYTE for supporting us here too! For detailed specifications of the system, those can be found below.
With the Ripjaws V now installed in the Intel-based system, we stopped off at the UEFI to enable the XMP 2.0 profile. We again get RAM at 3200 MHz with 16-18-18-38 2T timings, using 1.35 VDIMM. The board defaults to 1.312 VCCIO and 1.152 VCCSA.
When attempting to reduce the Intel rig's timings, we changed the VDIMM to 1.45V, left the VCCIO at 1.312V, but did change the VCCSA to 1.250V. Doing so allowed for the same results we got with the AMD system, with 14-17-17-38 but at a 2T command rate.
Using the same voltages as in the timings run, we rest the timings to match the XMP profile and added speed. We were stopped at 3466 MHz, which is another 133 MHz over our AMD results, which impresses us even more, as we through there would be no more headroom!
Chad's Intel DDR4 Dual-Channel Test System Specifications
- Motherboard: ASUS ROG Maximus XII Apex - Buy from Amazon
- CPU: Intel Core i7 10700K - Buy from Amazon
- Cooler: Corsair iCUE H150i Elite Capellix - Buy from Amazon
- Video Card: GIGABYTE GeForce RTX 2060 SUPER Gaming OC 8GB - Buy from Amazon
- Storage: Intel SSDPEK1W120GA
- Case: Thermaltake Core P3 - Buy from Amazon
- Power Supply: Corsair RM750 - Buy from Amazon
- OS: Microsoft Windows 10 Home 64-bit - Buy from Amazon
Using the DOCP profile, the 48145 MB/s result in read performance is a bit of a letdown overall. Lowering the timings only nets a 138 MB/s advantage, still behind the PNY and Thermaltake kits, but opting to run them at 3333 MHz delivered a 1953 MB/s advantage over DOCP and scores what we expected it should, compared to other kits, based on speed.
Again, behind the PNY and Thermaltake kits, but a wide margin this time, the DOCP profile offers up 44443 MB/s out of the box in write bandwidth. Opting to reduce the timings nets a huge advantage of 2082 MB/s over the DOCP run, and at 3333 MHz, the Ripjaws V slide into first with 3814 MB/s more than what we saw initially!
Copy performance is what we expect to see with the Ripjaws V's speed and timings! Beating both the PNY and TOUGHRAM using the DOCP profile, we have no complaints with the 49247 MB/s result. Running them at CAS14 gains 1562 MB/s, and at 3333 MHz, we gained 2402 MB/s but were not able to reach the XTREEM ARGB performance.
As we mentioned much earlier in the review, typically, with so many ICs at play, latency will take a hit. The trio of results in this chart shows that no matter how we set the kit to run, the results are lackluster compared to 16GB kits, but we are not surprised in the least to see latency play out like this.
We thought that the Super Pi results might have fared a bit better, but after multiple runs to verify scores, next to the last place is not where we had the Ripjaws V in our mind. Increasing speed is only marginally better, but reducing the primary and secondary timings puts the Ripjaws V in a range we expected out of the box.
3DMark Physics testing loves the density of the Ripjaws V! Right out of the box, they take top honors, and reduced timings results are even better than the DOCP results! Adding more speed to the kit did not do nearly as well, but can surpass the XTREEM ARGB.
The various testing methods of PCMark 10 hammers the Ripjaws V when used with the DOCP profile. Both of the overclocked results are more along the lines of what we expected to see, and both options fall between the PNY XLR8 and the TOPUGHRAM RGB.
Compressing lots of data is one of the things more density is good for, and the chart results prove this! Reduced timings win out over speed, but even the DOCP run leaves all of the Ripjaws V results in a solid second place!
Cinebench R15 is much like PCMark 10, with various tests to pass to deliver an overall score. The DOCP runs results are almost exactly what the PNY XLR8 delivered, but any attempt to overclock this kit delivers worse results than we initially saw using the DOCP setting.
Another thing that density plays a big part in increasing performance is when it comes to transcoding. Our Handbrake results show that the DOCP profile tops the chart without breaking a sweat! Lowering the timings is ever so slightly slower, but adding speed takes us down to the realm of the other 3200 MHz kits.
Read performance on our 10700K based system is impressive, to say the least! The 54330 MB/s XMP 2.0 result is 1504 MB/s better than the previous first place holder! While more speed or lower timings takes a hit in the scores, the results are still better than the other 3200 MHz kits in the chart.
Write performance is similar to what we saw in the PNY XLR8 RGB with a 46630 MB/s result. Running this RAM at CAS14 only nets a gain of 56 MB/s, but at 3466 MHz, there is a 4011 MB/s boost over what the XMP 2.0 profile delivered.
Better than both the TOUGHRAM RGB and the XLR8 RGB sets, the 43061 MB/s result is 1800 MB/s better than the competition. We gained 325 MB/s running the Ripjaws V at CAS14, but at 3466 MHz, we acquired 3026 MB/s over the XMP setting.
With more ICs comes more latency, and the results near the bottom of the chart were expected. While the overclocked runs did fare better, it is only slightly and is beat in this department by everything but the 2400 MHz Panther Rage.
Super Pi has the Ripjaws V in next to the last place, no matter which way we ran this RAM. Lowering the timings results in the slowest time to complete, with the XMP setting in the middle. Increasing speed shows better in the chart, but the difference to the XMP run is 0.1 seconds!
On the AMD system, Fire Strike loved this kit, but as you can see, on our Intel system, the XMP profile scores dead last! CAS14 results are where this kit should score using XMP in our minds, and even using them at 3466 MHz resulted in a score less than either the TOUGHRAM RGB or XLR8 RGB delivered.
The PCMark 10 scores line up better with what is expected at this speed on our Intel system. The XMP results fall between the other two 3200 MHz kits, and reducing the timings gets a slightly worse score for the effort. At 3466 MHz, we obtained the best results from the Ripjaws V and are just 32 points from the top spot.
7-zip compression bodes well for the Ripjaws V kit. The third-place finish for the XMP profile is decent, as it surpasses both the Thermaltake and PNY kits. The CAS14 run is 8.5 seconds faster than the XMP time, but the 3466 MHz run results in a fantastic time, some 21.5 seconds of your life back for every 7.62GB of data being compressed.
Cinebench R15 scores are similar to what we saw on the AMD rig, but this time the XMP results do not fall into last place, just close to it. Scores for either way to overclock this kit are very similar but still behind the other pair of 3200 MHz DDR4 in the chart.
Opposite of what we found with our AMD testing, all of the results are next to last on the chart on the Intel rig. While advantages are had with lowering the timings and a better advantage using more speed, it is tough to positively spin these results.
While a quick look at the charts will leave you with a feeling of mediocrity with the results these Ripjaws V were able to produce, outside of a select few benchmarks where they shined, you do not exactly buy a 64GB kit of DDR4 for outright raw performance that tops all of the charts! If you need this type of density, time is of the essence, as large laborious tasks ion productivity are better suited with reliability and stability first and foremost! Even if you need to wait a touch longer for the job to complete, the last thing you want is a system crash in the middle of a project. We can safely say that with 64 GB on these two sticks, running them under DOCP or XMP 2.0, we never ran into anything odd in the hours and hours of testing we go through.
While we do have to score the performance fairly, we also realize that we have moved away from "gaming" memory and have ventured into productivity, where the tuning of the kit is the key difference. How the Ripjaws V is laid out in manufacturing, we are pleased that we could run these dual-sided sticks, essentially a four-stick kit the way the motherboard sees it, on both AMD and Intel systems without a hiccup to mention!
On a more basic approach to looking at this kit, we get a lot of space to run for the average person getting into video production, 3D work, CAD, anything known to cripple a 16GB or 32GB kits under stress. However, these are not some generic naked sticks, either. While they do not offer any form of RGB, we do like the heat spreaders design, the use of black PCBs, and the bright white name on the top of them to show everyone what it is in the slots! While not "gaming" RAM by design, it plays the part where it dresses up the build with the appearance that we are already used to from G.Skill.
Considering the sheer amount of RAM we had in this pair of modules, to say we are not a little surprised that we could overclock them as far as we were able to would be a vast understatement. In past experience, when density increases, room to tinker diminishes. We had a lot of room with the timings, with only a tenth of a volt added to the VDIMM is interesting. On top of that, we were also able to reach 3333 MHz at 1T on the AMD system and 3466 MHz at 2T on the Intel system! Those with time to tinker with these Ripjaws V kits can deliver "gaming" RAM results, but it can be at the expense of stability and reliability.
Even when it comes to the price, we have no complaints. Being the most volatile part of any system, and the fact that these are designed for productivity, we feel that the $229.99 pricing is pretty darn good! Considering the money users spend on a CAD design system or video editing and the like, this is just one small drop in a much larger bucket! For those who need to obtain more density to alleviate bottlenecks in their daily workloads, we see no reason not to look at a kit such as this one!
The Bottom Line
The performance may not be spectacular, but this 64GB two-stick kit of Ripjaws V is still useful to the proper audience. We would not suggest this kit for the average gamer, but would certainly recommend it to the productivity crowd!