Jump to content
Search In
  • More options...
Find results that contain...
Find results in...


  • Content Count

  • Joined

  • Last visited

Everything posted by Firedrops

  1. Just got this set of RAM from Silicon Power and tried to tune it up, but there's essentially no information about it online and not even Thaiphoon Burner can identify the Die/Rev, just that Hynix is the manufacturer. So I'm making this thread and hoping it would be the central place to collect information and discuss it, and hopefully identify it better. Official Webpage: https://www.silicon-power.com/web/product-XPOWER_Turbine_RGB_DDR4_Gaming_Memory_ModuleHeatsink The ones I bought were 2x8GB Dual Channel, rated to 3200Mhz CL16. Thaiphoon Burner Report: MANUFACTURING DESCRIPTION Module Manufacturer: Silicon Power Module Part Number: SP008GXLZU320BSB DRAM Manufacturer: Hynix DRAM Components: H5AN8G8N??R-VKC DRAM Die Revision / Process Node: FFh / Not determined Module Manufacturing Date: Week 48, 2019 Manufacturing Date Decoded: November 25-29, 2019 Module Manufacturing Location: Taipei, Taiwan Module Serial Number: 00000000h Module PCB Revision: 00h PHYSICAL & LOGICAL ATTRIBUTES Fundamental Memory Class: DDR4 SDRAM Module Speed Grade: DDR4-2666 Base Module Type: UDIMM (133.35 mm) Module Capacity: 8 GB Reference Raw Card: A2 (8 layers) JEDEC Raw Card Designer: SK hynix Module Nominal Height: 31 < H <= 32 mm Module Thickness Maximum, Front: 1 < T <= 2 mm Module Thickness Maximum, Back: 1 < T <= 2 mm Number of DIMM Ranks: 1 Address Mapping from Edge Connector to DRAM: Standard DRAM Device Package: Standard Monolithic DRAM Device Package Type: 78-ball FBGA DRAM Device Die Count: Single die Signal Loading: Not specified Number of Column Addresses: 10 bits Number of Row Addresses: 16 bits Number of Bank Addresses: 2 bits (4 banks) Bank Group Addressing: 2 bits (4 groups) DRAM Device Width: 8 bits Programmed DRAM Density: 8 Gb Calculated DRAM Density: 8 Gb Number of DRAM components: 8 DRAM Page Size: 1 KB Primary Memory Bus Width: 64 bits Memory Bus Width Extension: 0 bits DRAM Post Package Repair: Not supported Soft Post Package Repair: Not supported DRAM TIMING PARAMETERS Fine Timebase: 0.001 ns Medium Timebase: 0.125 ns CAS Latencies Supported: 13T, 14T, 15T, 16T, 17T, 18T, 19T, 20T Minimum Clock Cycle Time (tCK min): 0.750 ns (1333.33 MHz) Maximum Clock Cycle Time (tCK max): 1.500 ns (666.67 MHz) CAS# Latency Time (tAA min): 14.125 ns RAS# to CAS# Delay Time (tRCD min): 14.125 ns Row Precharge Delay Time (tRP min): 14.125 ns Active to Precharge Delay Time (tRAS min): 32.125 ns Act to Act/Refresh Delay Time (tRC min): 45.500 ns Normal Refresh Recovery Delay Time (tRFC1 min): 350.000 ns 2x mode Refresh Recovery Delay Time (tRFC2 min): 260.000 ns 4x mode Refresh Recovery Delay Time (tRFC4 min): 160.000 ns Short Row Active to Row Active Delay (tRRD_S min): 2.799 ns Long Row Active to Row Active Delay (tRRD_L min): 4.525 ns Write Recovery Time (tWR min): 0.000 ns Short Write to Read Command Delay (tWTR_S min): 0.000 ns Long Write to Read Command Delay (tWTR_L min): 0.000 ns Long CAS to CAS Delay Time (tCCD_L min): 5.000 ns Four Active Windows Delay (tFAW min): 21.000 ns Maximum Active Window (tMAW): 8192*tREFI Maximum Activate Count (MAC): Unlimited MAC DRAM VDD 1.20 V operable/endurant: Yes/Yes THERMAL PARAMETERS Module Thermal Sensor: Not Incorporated SPD PROTOCOL SPD Revision: 1.1 SPD Bytes Total: 512 SPD Bytes Used: 384 SPD Checksum (Bytes 00h-7Dh): 6437h (OK) SPD Checksum (Bytes 80h-FDh): B9C3h (OK) PART NUMBER DETAILS JEDEC DIMM Label: 8GB 1Rx8 PC4-2666-UA2-11 Frequency CAS RCD RP RAS RC RRDS RRDL WR WTRS WTRL FAW 1333 MHz 20 19 19 43 61 4 6 0 0 0 28 1333 MHz 19 19 19 43 61 4 6 0 0 0 28 1200 MHz 18 17 17 39 55 4 6 0 0 0 26 1200 MHz 17 17 17 39 55 4 6 0 0 0 26 1067 MHz 16 16 16 35 49 3 5 0 0 0 23 933 MHz 15 14 14 30 43 3 5 0 0 0 20 933 MHz 14 14 14 30 43 3 5 0 0 0 20 800 MHz 13 12 12 26 37 3 4 0 0 0 17 INTEL EXTREME MEMORY PROFILES Profiles Revision: 2.0 Profile 1 (Certified) Enables: Yes Profile 2 (Extreme) Enables: No Profile 1 Channel Config: 2 DIMM/channel XMP PARAMETER PROFILE 1 PROFILE 2 Speed Grade: DDR4-3200 N/A DRAM Clock Frequency: 1600 MHz N/A Module VDD Voltage Level: 1.35 V N/A Minimum DRAM Cycle Time (tCK): 0.625 ns N/A CAS Latencies Supported: 16T N/A CAS Latency Time (tAA): 9.875 ns N/A RAS# to CAS# Delay Time (tRCD): 11.125 ns N/A Row Precharge Delay Time (tRP): 11.125 ns N/A Active to Precharge Delay Time (tRAS): 23.500 ns N/A Active to Active/Refresh Delay Time (tRC): 38.500 ns N/A Four Activate Window Delay Time (tFAW): 28.000 ns N/A Short Activate to Activate Delay Time (tRRD_S): 4.549 ns N/A Long Activate to Activate Delay Time (tRRD_L): 6.150 ns N/A Normal Refresh Recovery Delay Time (tRFC1): 350.000 ns N/A 2x mode Refresh Recovery Delay Time (tRFC2): 260.000 ns N/A 4x mode Refresh Recovery Delay Time (tRFC4): 160.000 ns N/A Show delays in clock cycles 1Usmus DRAM Calculator Screenshot: I am using a launch batch Ryzen 1600 (it even has the Linux kernel bug that I haven't bothered to RMA) on Asus Crosshair 6 Hero, 2 sets of these altogether so 4x8 GB = 32 GB. All these recommended values worked perfectly, boots on first try and so far no instability. I also checked the differences between selecting MFR vs AFR, and the only numerical difference was tRCDWR and tRCDWD swapping around (16 vs 17). I tried both set ups, and both boot with no discernible differences. I also ran the built-in benchmark to get these results: Unfortunately, without more information on the Die and Rev, trying to get FAST preset values just gives errors (not even values). So far, I'm EXTREMELY impressed by the timings these could reach on my CPU, where previous G.Skill TridentZ RGB 2x8 GB maxed out at 3200 cl 16-17-18-36 (was stuck at 17-19-19-40 until the 2019 bios updates) and wouldn't even boot any tighter, this works flawlessly despite being 4 sticks instead of just 2. I also got these for AUD$115 per pair, which is about USD$80. RGB B-dies typically cost almost double or more, so that's some insane value, too. Next few days I'll try decreaasing tRCDRD and TRP to 16.
  2. Hey, really appreciate the reply. Wanted to reply earlier but then did more non-rigorous tests/benchmarks and my conclusions kept changing... at some point I was pretty convinced it was not a performance issue, but "1.9x" and above was somehow special in software at some point in the chain, but then I tested a bit more and it seems this "1.9x" was only specific to anime for some reason. "IRL" tv shows can push much further, into ~2.xx speeds, even pretty high res/bitrate videos. However, there were also some old (low resolution and bitrate) shows that triggered it much earlier, at even ~1.4x. Maybe it has to do with the video's codec? Also, using radeon firmware to monitor CPU and GPU usage, they don't seem to hit 100% when the frame repeats begin. The highest I've got from video playback was ~15% CPU and ~80% GPU. In fact, video playback speeds seem to have very little to do with CPU/GPU usage on my system. It's still correlated, but quite little. IDK if it's because at really high playback speeds, my system realises it can't keep up so it just discards a bunch of frames from rendering? Or it could be because the graph smooths out spikes where the load hits 100%? Either way, appreciate the thought, looking forward to an updated guide and just let me know if you'd like me to run any specific tests
  3. So I followed this tutorial to get Potplayer properly set up with MadVR; Lately I've noticed that if I were to increase playback speeds to >1.8, frames would repeat. So the video will keep flashing frames from like a second ago, making it impossible to watch. However, some shows are so slow that 1.8x speed is not enough... Does anyone know how this could be fixed? Are there specific settings known to cause this, or overwhelms PCs? I'm using R5 1600, 16GB OC'ed ram, and 5700 XT, water cooled.
  4. I e-mailed Asus to ask about it, they gave me back a canned response. "No Plans for releasing the ROG CROSSHAIR VI HERO (WI-FI AC). At the moment the non-wifi version is available." No plans at all???
  5. Yup, according to google translate. Who knows what's going on with pricing ¯\_(ツ)_/¯
  6. I wish they would just release the damn board already, it's almost a month from when they announced it. That Swedish retailer doesn't have it in stock either, it's just a pre-order. I have all my parts ready except for the motherboard. Nobody sells the 7260/7265 with the internal wiring and external antenna here, just the bare card. UGH.
  7. Jesus, how did those guys get 130+?!
  8. YES! Because... YES! That's all. . . . . . YES!
  9. My favourite feature is the front-facing stereo speakers. I'm really with Linus on this one about how nobody else cares to implement such an obvious "feature", and they've already been given a year to copy.
  10. IMO it's a close tie between CM's HAF XM and Fractal Define, in my experience they offer the best for mid-towers.
  11. Hold up! NZXT seems to have released a "successor" for the Switch 810's price point, the phantom 630. Do check it out!
  12. As the others have posted, those can do quite well for E-ATX, though I wouldn't recommend the 410/600t over the HAF-XM, especially specifically for E-ATX!
  13. Personally I'm leaning quite heavily towards the NZXT 630. About the size, it is actually very efficient with its space. A tad smaller than the 820, but still offers XL-ATX mobo support. It looks amazing(personal preference, really) VERY modular and customizable(2 SSD mounts behind the mobo tray,some of us won't even need any HD cages at all) great potential for water cooling (easily hold about 5 x 140mm rads, maybe a few 200mms with a little modding) Very neatly holds whatever mobo due to its recessed mobo tray design. Built-in fan controller, SD card reader, and I/O LED Anandtech's review even showed that at stock, this case defeats the 820 in cooling. Some other sites already posted unboxing vids on youtube, but I'm still waiting for Linus', he's the best. Just my 2c. Good luck and do let us know which case you pick!