Since me and you share a similar issue, can you provide a link to newegg for the exact ram package you have?
K, this is the very stick I would have bought if I didn't get the RGBs.
I was planning on doing this myself. Specifically, after getting VRM waterblocks, I was going to either find pre-made ones or buy a copper sheet, drill holes where appropriate so I can use thermal pads with the backside components to the VRM, make sure for clearance, use electrical tape (I have both tape and liquid to spread in a sheet; meant to prevent electrical shorts by contacting traces and solder points on the back), but was considering just leaving a larger part of the plate to sink the back of the socket. My mount for the Core X9 cube has where the drive part is lower, allowing a wind sweep under. So I can attach heatsinks to the plate in the area of that recess and use the fans in the side of my case to blow air in to cool it in part, or so my thoughts go. But, since I'm at my desktop, screenshot time!Question : Has anyone ever thought of or does anyone here put a fan on the backside of their mobo in the area where you have a mobo cutoff?
I noticed a bit of activity w.r.t to chips back there and I notice that the backplate gets quite toasty when you're pushing the system and figured it wouldn't hurt. I've also noted from reviewers with FLIR cameras that things can get kind of toasty back there. Thoughts?
I know. I have a lot more work to do on that score (I take it this is the same person I know from NBR and 0_o from HWBot). This is more me just messing around at the moment. I have A LOT more to do after I buy the custom loop later this month after I finish this project in about a week and a half. I didn't even start up in diagnostic mode for this, although I did set high performance in priority, IIRC. That is one of the scores that has been bugging me though, although the 1024m is looking nice. I also just moved to the fall creator's edition beta, which evidently just hates wPrime (lower scores on both). Also, realbench pops up saying that the build has the RTC bug, so I cannot submit the newer scores since I updated to that, as anything without HPET enabled would then violate the rules (I don't get the warning on the Spring Creator's edition, though, so they likely resolved it on that build). That's also why I didn't post the CB11.5 where it had 27.2, I believe. If that was caused by the RTC bug, no use in sharing. I got 22.2 on it on the Spring Creator's edition, though. But more will be coming once I have time to play!@ajc9988 that 32M wPrime run seems longer than I would have expected. Since it runs really short on these processors did you try running with high performance? ie get that multi up to max before the start rather than stepping the multi up through the bench.
Besides benchmarking, this CPU is such a beast once you start throwing heavy workloads at it. Already have 2 VMs running with a few Visual Studio agents compiling code and it eats through anything you throw at it. Can still game while it's compiling code in the background with no FPS drops. Amazing.
Keep in mind that there have been problems with the RGB because of how it siphons power for the lights. I know early Ryzen 7 systems were exploding these chips left and right.K, this is the very stick I would have bought if I didn't get the RGBs.
So, we've ruled out RGB having less quality as a cause as you are having the issue w/ non RGB.
I thought that the problem was the beta nature of the software for controlling the RGB LEDs on them, and that it would screw up (especially if you were running other hardware monitoring software), and accidentally re-program and corrupt the SPD on the DIMMs, after which, the mobo didn't know what to do with them on next cold boot.Keep in mind that there have been problems with the RGB because of how it siphons power for the lights. I know early Ryzen 7 systems were exploding these chips left and right.
You are probably right I just remembered that people were "killing" their chips a lot when I was building my R7 1700 system. Since I was going to spend that amount for shiny lights I didn't really track I much after that.I thought that the problem was the beta nature of the software for controlling the RGB LEDs on them, and that it would screw up (especially if you were running other hardware monitoring software), and accidentally re-program and corrupt the SPD on the DIMMs, after which, the mobo didn't know what to do with them on next cold boot.
To the naive, it would look like the DIMMs were fried, but really, they just needed their SPD re-programmed.
Phanteks have a waterblock for TR4
and will add support for TR4 to their dual-tower air cooler PH-TC14PE:
Phanteks have a waterblock for TR4
and will add support for TR4 to their dual-tower air cooler PH-TC14PE:
@The Stilt, TC14PE's base is not full cover indeed, but it seems to me it provides a bit more coverage than Asetek's base. (I might be mistaken though.) Furthermore, I read in an older review that its base is planar, not convex like some (or all?) of the Asetek based AIOs.
Thanks. Which is the longer of the two dimensions, in parallel with or perpendicular to the heat pipes? In parallel would be bad, because then the overhang of the dies would be even worse.TC14PE base is 42.5x45mm, which is even smaller than the Asetek AIO's (52x52mm).
You have :So whats the latest on the coolers? Any comprehensive reviews of the fully sized TR4 coolers vs. the compromised options @launch?
Yeah that contact plate size looks like a joke... Noctua did this better.Thanks. Which is the longer of the two dimensions, in parallel with or perpendicular to the heat pipes? In parallel would be bad, because then the overhang of the dies would be even worse.
Their water block looks like it has the best fin surface I've seen. It really looks like a beast :Nice. Didn't know phanteks came out with a TR block. XSPC just released theirs and swiftech started taking pre-orders for their apogee TR waterblock. Heatkiller should be next as well.