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

PCI-E 4.0 SSD tested - slower than 3.0

Joined
Aug 6, 2017
Messages
4,493 (6.06/day)
Location
Poland
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) RTX 2070 Super Gaming X Trio
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480,M9Pe(Y) 512 (games)/4TB HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) W830BT headphones
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
seems like with the exception of synthetic scores the drives are performing much better on intel's pci-e 3.0
if you take a look at small file copy,multitasking (simultaneous copy+install) and installation times, the same drive is 15% faster on intel's 3.0 than ryzen's 4.0
LINK TO FULL REVIEW

damn that mp600 is fast though.
127446
127447
127448
 
Last edited:

Aquinus

Resident Wat-man
Joined
Jan 28, 2012
Messages
11,290 (4.09/day)
Location
Concord, NH
System Name Kratos
Processor Intel Core i7 3930k @ 4.5Ghz
Motherboard ASUS P9X79 Deluxe
Cooling Corsair H100i V2
Memory G.Skill DDR3-2133, 16gb (4x4gb) @ 9-11-10-28-108-1T 1.65v
Video Card(s) Sapphire AMD Radeon RX Vega 64
Storage 2x120Gb SATA3 SSD Raid-0, 4x1Tb RAID-5, 1x500GB, 1x512GB Samsung 960 Pro NVMe
Display(s) 1x LG 27UD69P (4k), 2x Dell S2340M (1080p)
Case Antec 1200
Audio Device(s) Onboard Realtek® ALC898 8-Channel High Definition Audio
Power Supply Seasonic 1000-watt 80 PLUS Platinum
Mouse Logitech G602
Keyboard Rosewill RK-9100
Software Ubuntu 18.04 (5.2.3 Mainline Kernel)
Benchmark Scores Benchmarks aren't everything.
Or maybe single-threaded performance is the bottleneck and not the drive? Installation could be doing a lot more than just copying data. For example, the installation data might be compressed, in which case you're testing the decompression algorithm more than I/O. They needed other 3.0 devices tested on the same AMD platform in order to come to the conclusion that PCIe 4.0 is actually slower (which it's not, it's far more likely that latency is the problem, which would be apparent on PCIe 3.0 devices as well,) when that's contradicted by the synthetic benchmarks which are designed to specifically only test I/O.
 
Joined
Nov 11, 2004
Messages
3,652 (0.68/day)
Location
Formosa
System Name Overlord Mk MX
Processor AMD Ryzen 7 3800X
Motherboard X570 Aorus Master
Cooling Corsair H115i Pro
Memory Viper Steel DDR4 3733MHz 16-19-19
Video Card(s) Gigabyte RTX 2080 Gaming OC 8G
Storage 1TB WD Black NVMe (2018), 2TB Viper VPN100, 1TB WD Blue 3D NAND
Display(s) Asus PG27AQ
Case Corsair Carbide 275Q
Power Supply Corsair RM750
Mouse Logitech G500s
Keyboard Wooting Two
Software Windows 10
Benchmark Scores https://valid.x86.fr/ztiub6
1. Expand your sources.
2. You could start here
3. Keep in mind that the Phison E16 controller is a PCIe 3.0 controller strapped to a PCIe 4.0 bus, it's not a good representation of what PCIe 4.0 SSDs will be capable of.
 
Joined
Aug 6, 2017
Messages
4,493 (6.06/day)
Location
Poland
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) RTX 2070 Super Gaming X Trio
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480,M9Pe(Y) 512 (games)/4TB HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) W830BT headphones
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
1. Expand your sources.
2. You could start here
3. Keep in mind that the Phison E16 controller is a PCIe 3.0 controller strapped to a PCIe 4.0 bus, it's not a good representation of what PCIe 4.0 SSDs will be capable of.
sorry,but no one except ppc tested intel's 3.0 vs amd's 4.0
neither the link you provided
Or maybe single-threaded performance is the bottleneck and not the drive? Installation could be doing a lot more than just copying data. For example, the installation data might be compressed, in which case you're testing the decompression algorithm more than I/O.
well not for file copy probably.
and doesn't amd boast highest ipc ?

btw this is pretty much in line with what ryzen 1000 showed


so no improvement for 3 generations despite the 4.0 feature
 
Last edited:
Joined
Jun 18, 2010
Messages
2,017 (0.60/day)
Processor Intel i7 970 // Intel i7 2600K
Motherboard Asus Rampage III Formula // Asus P8P67 Deluxe
Cooling Zalman CNPS9900MaxB // Zalman CNPS11X
Memory GSkill 2133 12GB // Corsair V 2400 32GB
Video Card(s) ASUS GTX1080 // MSI GTX1070
Storage Samsung 860EVO // Samsung840P
Display(s) HP w2207h
Case CoolerMaster Stacker 830se // Lian Li PC-9F
Audio Device(s) onboard
Power Supply Seasonic X 850w Gold // EVGA 850w G2
Mouse Logitech G9
Keyboard Dell
Software W10 Pro 1903
@cucker tarlson : Did you see and read this yet?

 
Joined
Aug 6, 2017
Messages
4,493 (6.06/day)
Location
Poland
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) RTX 2070 Super Gaming X Trio
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480,M9Pe(Y) 512 (games)/4TB HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) W830BT headphones
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
Joined
Jan 4, 2013
Messages
171 (0.07/day)
Location
Denmark
System Name R7 3800x@4525MHz/E5800@4GHz/Skylake 6400
Processor R7 3800x/E5800/i5 6400
Motherboard Gigabyte Master X570/Asus P5QL-EM/Asus Z170 Pro Gaming
Cooling Noctua NH15D-SE/Zalman 9500 series/Stock
Memory 16GB CMK16GX4M2B3600C18/Kingston HyperX PC8500 - 4GB/G.Skill Ripjaws 32 GB F4-3200C16D-32GV
Video Card(s) R7/HD7800XT / R290X
Storage Aorus 2TB Corsair MP600 + 960 Evo 1TB/Seagate 7200 series 12, 11, 10 and 9 / 2xSamsung EVO 850 500GB
Display(s) 2x Acer P243w and 1 Acer G245h in the midle
Case Custom builds
Audio Device(s) -
Power Supply Silverstone 850watt modular Gold/Gold Silverstone 850watt/1000Watt Antec
Software win10pro / Win10 Home / win7 / wista 64 bit and XPpro
Well it depends on test methods - I had a much higher score on my old Z170 based system than my x570, but In real life I am pretty sure that they score of my x570 is the more realistic one.

127452

Above from my Skylake Z170 setup

127453

Same disk on my x570 setup

So its not easy to compare these benchmark
 
Joined
Aug 6, 2017
Messages
4,493 (6.06/day)
Location
Poland
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) RTX 2070 Super Gaming X Trio
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480,M9Pe(Y) 512 (games)/4TB HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) W830BT headphones
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
lol that is not possible to have 8000mb/s sequential read and 250mb/s 4k random read.
btw the review is comparing z370 and 8700k,I think older gens (kay and older) were much more heavily impacted by smeltdown patches.On my 5775c/z97 4k write dropped from 110mb/s to just 85mb/s (-23%) after the patch,disabled it's back to normal.That's probably why TPU shows sx8200 running on 7700k at 93% performance of ryzen.Wonder why test Kaby.
 
Joined
May 7, 2018
Messages
1,167 (2.49/day)
Location
United Kingdom, London
System Name (PC / PS4 / PS3 / PS2 / XBOX 360) and other stuff
Processor Sorting out the CPU / 6600k 4.9ghz 1.43v/x5650
Motherboard X299 Omega Extreme/ Z270X gaming 7/Clevo x7200
Cooling alphacool eisbaer 360mm/Silent loop 280mm/Stock
Memory 32gb ddr4 corsair vengance lpx / None yet/24gb ddr3 1333mhz
Video Card(s) Evga gtx 1080 ti ftw3 @ 2000mhz/5508mhz XOC BIOS + alphacool nexxos gpx / asus 560 ti 1gb/ 460M sli
Storage kingston ssdnow uv400 480gb ssd (os) toshiba x300 6tb (everything else)/ 750gb/10tb Elements
Display(s) ElectriQ 4k 28" TN 60hz / Samsung 931cw /1080p 17.3" laptop
Case Corsair Carbide 600c/Be quiet! Dark base 700/Lego pc case/clevo x7200
Audio Device(s) Hyper x cloud revolver s
Power Supply Seasonic snow silent 750w/Be quiet power pro 11 750w
Mouse Mad Catz Rat Pro X (x4, one with broken off scroll wheel) / Lynx bluetooth controllers x3
Keyboard Havit gaming keyboard (amazon keyboard+mouse bundle)
Software Windows 10 pro + Kali linux
Benchmark Scores Too many to fit here.
Same disk on my x570 setup

So its not easy to compare these benchmark
It is, if you notice you've actually ramcached the drive on Z170.
 
Joined
Jan 4, 2013
Messages
171 (0.07/day)
Location
Denmark
System Name R7 3800x@4525MHz/E5800@4GHz/Skylake 6400
Processor R7 3800x/E5800/i5 6400
Motherboard Gigabyte Master X570/Asus P5QL-EM/Asus Z170 Pro Gaming
Cooling Noctua NH15D-SE/Zalman 9500 series/Stock
Memory 16GB CMK16GX4M2B3600C18/Kingston HyperX PC8500 - 4GB/G.Skill Ripjaws 32 GB F4-3200C16D-32GV
Video Card(s) R7/HD7800XT / R290X
Storage Aorus 2TB Corsair MP600 + 960 Evo 1TB/Seagate 7200 series 12, 11, 10 and 9 / 2xSamsung EVO 850 500GB
Display(s) 2x Acer P243w and 1 Acer G245h in the midle
Case Custom builds
Audio Device(s) -
Power Supply Silverstone 850watt modular Gold/Gold Silverstone 850watt/1000Watt Antec
Software win10pro / Win10 Home / win7 / wista 64 bit and XPpro
It is, if you notice you've actually ramcached the drive on Z170.
My rapid mode and Asus ramcache where actually attached to my two SATA SSD drives so thats not it - but for some strange reason the score with both my Samsung EVO 960 and the Corsair MP600 where high. So there must be some other windows/intel caching that helped the score. The point being that its is dificult to compare these benchmarks across different platforms
 
Joined
Aug 6, 2017
Messages
4,493 (6.06/day)
Location
Poland
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) RTX 2070 Super Gaming X Trio
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480,M9Pe(Y) 512 (games)/4TB HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) W830BT headphones
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
The point being that its is dificult to compare these benchmarks across different platforms
no,the higher number wins.
 
Top