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


  • Content Count

  • Joined

  • Last visited


This user doesn't have any awards

About MageTank

Contact Methods

  • Battle.net

Profile Information

  • Location
    United States, Ohio
  • Gender
  • Interests
    Gaming, Computer Hardware
  • Occupation
    Slim Jim Enthusiast


  • CPU
    Core i7 8700k 5.4ghz Cinebench Stable (best kind of stable)
  • Motherboard
    ASRock Z370 Fatality K6
  • RAM
    4x8GB Patriot Viper Steel DDR4 4400 C19 (Clocked at 4000 C15-15-15-30-2, 36ns latency)
  • GPU
    EVGA RTX 2080 Ti Black Edition XC
  • Case
    Thermaltake Core P3
  • Storage
    Intel 2TB 660P M.2 NVMe SSD
  • PSU
    EVGA 850W Supernova G2
  • Display(s)
    LG OLED B9 55" 4k 120hz G-Sync TV
  • Cooling
    Decent Sized Custom Loop
  • Keyboard
    Logitech G Pro TKL
  • Mouse
    Logitech G703
  • Sound
    Sennheiser Game One
  • Operating System
    Windows 10 Pro
  • Laptop
    PowerSpec 1510 (Clevo P650HS-G) w/ 120hz G-Sync panel

Recent Profile Visitors

12,490 profile views
  1. I plan to jump ship from Intel Z370 to AMD's X570 platform for a 5950X. Looking for some X570 board recommendations. My primary focus is memory overclocking above all else. CPU overclocking will be toyed with, but will likely let PBO handle that as it has consistently out-performed what I am capable of in my testing.


    My job has granted me access to the ASRock X570 Taichi and X570 TUF Gaming from ASUS, however I am not a fan of either of those boards in particular. I've been looking at the ASUS DTX Impact board as well as their X570 ITX ROG Strix board, however I have always been annoyed with the ASUS BIOS given their weird memory scrambler features and "memory modes" that I can't seem to ever disable. If anyone has any other recommendations I should keep an eye out for, I'd appreciate it.


    Other things to note, I dislike RGB and I prefer neutral color tones as the theme of my build will be entirely stainless steel with copper hard tubing.

    1. Show previous comments  6 more
    2. Jurrunio


      Btw I think you already know about Zen 3's bigger than usual gains when you get more ranks of memory into each channel?

    3. MageTank


      I've had my hands on Zen 3 CPU's for quite a while now. It's no different than Zen 2 as far as behavior with memory ranks are concerned, people have been misreporting those gains mostly due to an error in their testing methodology. 2x16GB (2xDR for 4 ranks total) is still the best configuration in my testing. Anything beyond that is subjected to reduced performance due to 2DPC IMC stress penalties induced on clock speeds/latency and 4x1SR still lacks rank interleaving.


      I plan to run my current memory kit with most likely the B550 Unify X, with the goal of achieving DDR4 4000 with C15/C16 timings: https://www.gskill.com/product/165/166/1594279359/F4-4266C17D-32GVKBRipjaws-VDDR4-4266MHz-CL17-18-18-38-1.50V32GB-(2x16GB)

    4. MageTank


      This B550 Unify X board doesn't appear to be launching in any stores any time soon. Seems to be a paper launch in every sense of the word. I've been sitting on a 5950x for a while now, hopefully they launch this board soon or I may need to find alternative options.

  2. My point was that the original analogy of selling a 10M home for 5M home and being unable to call the seller "rich" was weak based solely on the premise that selling a home for a 5M loss automatically prohibits the use of that description. I am sure the point you were making in and of itself was valid, I just took issue with the analogy you chose to use. That's all.
  3. Sounds like the SBA found the necessary collateral required to make this loan. Like I said, most tech startups receive investments, not loans. This one happened to get a loan after investors already got on board, which is what allowed them the necessary collateral required to secure the loan in the first place. Lenders are not stupid. They are not in the business of giving their money away to failing endeavors, not without the means to recoup most if not all of their losses entirely (and in most cases, with guaranteed interest). Your own source leans more in agreeance to my point o
  4. Sadly my only alternatives at this location is satellite internet, and I just don't know if I trust the idea of satellites from a latency perspective. The current owners have Frontier and claimed they had no issues, however they are much older and do not strike me as the "power user" type. I suppose if things don't work out well, we go with plan B, and run an extremely long CAT6 cable to my nearest neighbor with Spectrum and ask if I can borrow a cup of gigabit, lol. I don't know. It's entirely possible that multiple people missed the point of your analogy, but I still feel it's
  5. Yeah, mostly because the analogy is weak. It operates on the assumption that the person that sold the home is automatically in debt because they sold it for a loss. It also operates on the assumption that they are somehow poor despite somehow convincing a bank to approve a loan for a 10 million dollar home, which requires that they had a solid income or equity to put up against the loan in the first place. If your entire analogy hinges on the premise that the person that sold their $10m home also lost their high paying job, the valuable assets they possessed that aided in the securing of their
  6. Meanwhile here in the States, I will be moving from the city to the country in a couple weeks, going from gigabit down to bonded ADSL. The sad part is, my new land will be surrounded by gigabit offerings just 2 miles around me, just my land is a deadzone with no hub.
  7. Sounds like some of the older standard B-Die, all single rank. I am still curious as to what he means by "select secondary and tertiary timings manually controlled" as depending on the selection, those are very important. For example: If he controlled SG/DG (Same Group / Different Group) timings but not DD (Different DIMM) timings (DR or Different Rank wouldn't matter for his configuration given the use of single rank DIMM's), he would be dealing with an issue where the board is training his DPC timings which would definitely skew the results of his testing. I'd imagine Steve is smart enough t
  8. If he is going through and setting every single secondary and tertiary timing, I applaud his effort, but that is absolute madness. Unless he is doing so from a median perspective (an overclock configuration that all of his CPU's can achieve), this in and of itself would introduce more variables for his testing if memory controller quality differs per chip. This is extra troublesome on modern boards that let you POST with obvious unstable RAM overclocks but end up correcting the instability through additional training/correction procedures that negatively impact performance. If you have any inf
  9. I have 2 of each Zen 5000 series CPU currently out right now in my test lab along with almost every possible memory kit you could imagine. I've driven myself a little crazy the past few weeks with testing and have not observed what Steve is claiming Wendell observed. What I observed is a direct contradiction. 2x16GB is consistently out-performing the other configurations, followed by 2x8GB single rank (clock speed advantage, rank disadvantage), followed by 4x8GB (rank advantage when interleaving channels, clock disadvantage due to 2DPC timing configurations, especially when using d
  10. I don't think it's much of a mystery at all. We've always seen these kinds of interactions on memory controllers, even spanning back on older Intel platforms. I also don't believe it has anything to do with single rank DIMM's in particular as Steve didn't test dual rank DIMM's in his reviews. There are some important things to note involving this subject. #1. When you populate 4 DIMM's, you activate 2DPC (DIMMS Per Channel) timings that are otherwise completely disabled or not in use. These timings, while technically more stressful on the memory controller, do have an
  11. Can't really consider that an apples to apples comparison though. Not too many attachments one can attach to toilet paper. If a store offered me a warranty or replacement plan for my toilet paper, I'd be pretty concerned... lol.
  12. No, but retailers do care about money from attachments. They'd rather sell to genuine customers that would also be willing to buy replacement plans and other components alongside a scarcely available product instead of selling an item in-bulk at MSRP that is just going to flip it in return for a higher price. I do vividly recall the shortage of GTX 1000 series cards due to crypto mining and retailers that were offering cards at MSRP for people that were buying other system components alongside the cards. I think it is in the best interest of these retailers to reach as
  13. Nobody is arguing that it's bad from a business decision perspective, if AMD wants to game the system to make a buck and it is legal for them to do so, I say go for it. My only point is that their intent behind doing so can be seen as potentially predatory/misleading for consumers as it is designed to intentionally confuse them based on how extremely similar the naming conventions are. I am not faulting AMD here as the sole party responsible for these practices as plenty of companies do it world wide, not just in the tech industry either. It's just my opinion that AMD's track record is the wor
  14. I don't think AMD had to name it that similarly to Intel though. They could have called their chipset the AMD X300 back when Intel had the Z270. 300 is still a larger number than 270, if that was their plan the entire time. X399 vs X299 was by far the biggest slap in the face seeing as both had the same starting letter as well, but luckily that was comparing two enthusiast platforms and I'd hope that enthusiasts were harder to confuse. For actual CPU naming, I don't really have a problem with AMD's adoption of the R3, R5, R7 nomenclature in the CPU segment at all, especially consid
  15. I'll be bold enough to say we can go ahead and remove the "might" from your statement and double down on them absolutely being worse with their naming conventions. AMD is intentional with their naming. They do it on purpose with the intent to confuse the consumers into believing their product is superior (Which is weird, because all signs are pointing to this being true without the shenanigans being required). Intel releases Z270, AMD releases X370. Intel releases X299, AMD releases X399. I can't imagine how many poor retail workers had to deal with explaining the difference between X370 and Z