• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Kingston Fury Renegade G5 2 TB

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,980 (3.75/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
With the Renegade G5, Kingston expands their solid-state drive offerings to finally include PCI-Express 5.0. Their new drive is built using the Silicon Motion SM2508 controller, paired with Kioxia TLC NAND flash, which offers a better performance profile than the Micron TLC that we've seen on similar drives.

Show full review
 
@W1zzard this on Page 20 needs to be fixed as well ("GB" needs to be "TB"):

1749836764739.png
 
The most important thing most (gamers) should know:
  • In real-life not significantly faster than the best Gen 4 drives
No need to pay 2x the price for PCIe 5.0.

And I am still waiting for an affordable 8TB PCIe 4.0 NVME.
In germany the cheapest one is the SN850X with a price of 72€/TB, while 2TB drives are under 50€/TB...
In 2023 I bought my Samsung 870 8TB QVO for 300€, that is under 40€/TB.
Sadly the SSD cartel keep the prices high and raise them even.
 
Wow does that use a lotta power...
Efficient but dang.
 
Buy a Lexar NM790 or Silicon Power XS70, call it a day and don't torture yourself with these PCIe 5.0 under engineered "marvels".
 
Buy a Lexar NM790 or Silicon Power XS70, call it a day and don't torture yourself with these PCIe 5.0 under engineered "marvels".
I'd rather buy another Gen 3 drive, these things just get too hot.
 
The most important thing most (gamers) should know:
  • In real-life not significantly faster than the best Gen 4 drives
No need to pay 2x the price for PCIe 5.0.

And I am still waiting for an affordable 8TB PCIe 4.0 NVME.
In germany the cheapest one is the SN850X with a price of 72€/TB, while 2TB drives are under 50€/TB...
In 2023 I bought my Samsung 870 8TB QVO for 300€, that is under 40€/TB.
Sadly the SSD cartel keep the prices high and raise them even.
This thing is, some of us don't want a super slow shitty QLC drive, yes you got an 8TB version but QLC absolutely doesn't work in my use case, I've used MSI's M570 pro and Samsung's 9100 Gen 5 drives between the 2 no real difference, I also have KC3000's (which for being a gen 4 drive sometimes beat some Gen 5 drives) and a few WD SN850x. I need fast sustained transfers and QLC drives while cheap have terrible sustained transfers when dealing with large file transfers,

I have my Gen 5 drives as the boot/os/programs and Raid 0 KC3000's because I need fast file copies.

Do I think Gen 5 drives are expensive? They absolutely are and not the best deal but in some cases it's what people want and need.
 
Wow does that use a lotta power...
Efficient but dang.
Except that for huge majority of users efficiency under load matters little.
For them drive propably doesn't even spend 1% of total time doing something counting as actual load for the capabilities of NVMe.

Without "spinning rust" in there less than 1W should be enough for idling.
And even power saving modes don't seem to help much.
Laptop manufacturers are soon going to have major headache in finding drives not sucking power for doing nothing.
 
So there is no need to buy a Gen 5 SSD for 99% of customers. In real life the best Gen 5 SSD is only 10% faster than Gen 4 SSD but so much expensive, running hot and consumes more power. So, i prefer to buy 4 TB Gen 4 SSD over 2 TB Gen 5 SSD.
 
Analyzing how the large file copy test was performed, there's at least one thing that seems suspicious to me. It's explained that 75 GB was copied to another disk. I'd like to know which disk it was copied to, since the write speed of the other disc, directly influences the speed being measured. It probably wasn't copied to another same disk because the Asus ROG MAXIMUS Z790 Dark Hero motherboard only supports one M.2 Gen 5 disk.
 
Unlike the GM9000 and some other competing models, the Fury Renegade G5 is built with Kioxia 218-layer BiCS8 TLC NAND flash instead of Micron's B58R.
You mean this week.
 
In a complete standstill in PC storage they have now announced PCIe 6 won't arrive until 2030, although specification has been finalised for some time.

PCIe 6.0 SSDs for PCs won't arrive until 2030 — costs and complexity mean PCIe 5.0 SSDs are here to stay for some time

And there have been absolutely no news on larger capacity 3D NAND chips for consumer drives that were promised by all the makers several times now, arriving "soon" - I guess large enterprise orders for LLM, AI made them totally abandon their plans for consumer sector.
 
Yes - just marketing. No performance gain. Kingston KC3000 2TB or similar PCIE 4.0 drivers wiht proper DRAM are performing quite similar. Very sad. Same 2TB and basically the same performance. Slight better sometimes but that depends on the measurement scenario.
 
I have 2 of these in their 4TB SKUs... They are lovely. No thermal throttling in my case as they don't get loaded that heavily and the ProArt X870E has decent enough heatsinks and thermal pads.
really expensive but thats what we have to deal with to get speeds like this right now
 
I see that in most disk tests, they all seem to perform the same. I have to say they're poorly tested. In real life, there's a huge difference between them. For example: take a Steam game like Call of Duty and verify file integrity. Or take a game like GTA V and move it to another partition on the same disk. There can be differences of up to an hour.
 
I see that in most disk tests, they all seem to perform the same. I have to say they're poorly tested. In real life, there's a huge difference between them. For example: take a Steam game like Call of Duty and verify file integrity. Or take a game like GTA V and move it to another partition on the same disk. There can be differences of up to an hour.
downsides of controlled synthetic testing and benchmarks... IRL will always be different, due to non-linear control in most cases
 
I see that in most disk tests, they all seem to perform the same. I have to say they're poorly tested. In real life, there's a huge difference between them. For example: take a Steam game like Call of Duty and verify file integrity. Or take a game like GTA V and move it to another partition on the same disk. There can be differences of up to an hour.

1000005122.jpg



Copying between partitions is a very specific use case that tells you exactly how the drive performs this, and doesn't tell you much about performances in other uses. Simultaneous continuous reading and writing of the same data on the dame drive can really expose that some drives don't handle that well.

But I haven't made multiple partitions on a single drive for decades. So I don't care, and I think a test that would put a lot of weight on that performance would be useless for my uses of the drives.

The truth is that majority of the SSD drives from PCIe 2.0 onwards perform about the same in most real world uses. Installation speeds, game and app load times etc. Sure, you can find specific cases where some drives perform better, or some drives particularly worse. But I think drive manufacturers would welcome any such test that they could use to show their drive really is faster in use, not just in meaningless synthetic tests (measuring consumer drive speeds in high queue depths)... But noone will care if it's oddly specific and not consistent with other use cases.
 
Back
Top