They did this in practice anyway, they just updated the text.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Ryzen "2" ?
- Thread starter beany_bot
- Start date
They did this in practice anyway, they just updated the text.
Ye but they could say Sorry mate not covered kinda
Ye but they could say Sorry mate not covered kinda
In theory, but in practice it was to discourage people from taking off the stock cooler and putting on something worse. Doing that will force the cpu to perform under spec if it overheats and this must have happened at some point.
So to head off the idiots they wrote it in.
Thats not the point. My rule of overclocking is. EVERYTHING needs to work. I could get better clocks possibly by going full out constant 1.43-1.45 no downclocking or power daving ect. But where is fun in that?? Thats the easy mode.
Try having high overclock on everything stable and everything still working. Can take 50 or 100 hours of tuning. But then satusfaction is better
Besides that I dont use **** windows task power plan. I use Proces lasso for years so.much better cpu management
Not like i care about power saving after all its inder 100% load 24/7 yesterday had longest rest in months when i was swapping motherboards.
Atm its runing hci memtest on new test seting infinite loop so I will see how far without errors can it go
New vrm design on soc and ddr4 with shielded ports could give me cl14 finally. Atm 1.44 ddr and 1.125 soc test.
Last edited:
JayzTwoCents saw the default voltage of his chip at stock was over 1.45 V.Trying it now with -0.1 and PE level 3. All core clock goes slightly higher(4050 ish) but max boost slightly lower (4350).
All core ends up about 1.2v and single core boost does hit 1.45, think that's fine on these processors.
Seems these can be pushed quite hard with bclk increase, bit more offset voltage and PE, 4.5+ boost clocks, not sure its worth it though.
Thanks, will watch that over lunch. Definitely an interesting test because it shows what the potential of AMD's Zen architecture is if they can get the clocks higher with 7 nm. One would assume they'll get bigger IPC improvements with Zen 2 but you never know: it could be that they moved engineers over to Zen 2 pretty early, which is why Zen+ had very minimal IPC gains (just from memory latency improvements), or it could be that they are just finding it hard to improve IPC in general. I'm hoping the former is true but it'll be a while before we know.
Permabanned
- Joined
- 12 Sep 2013
- Posts
- 9,221
- Location
- Knowhere
I got around to replacing my 2700x's stock cooler with a H100i GTX AIO yesterday, I was unable to do certain stress tests like Prime 95, Timespy's extreme & Firestike's extreme with the memory at 3200mhz 14.14.14.31, They'd all crash within a couple of minutes, so I dropped the memory speed to 3000mhz with the same timings and it no longer crashes. I haven't tried overclocking it yet and probably won't bother unless we start seeing gaming increases that make it worth while. As it is all I need is the memory sorted & then I'm happy with it as is.
With the Prime 95 the H100i temps maxed at 60 degrees which is good compared to the mid to high 70's the stock cooler did yesterday, that said it was only a 10 minute test so I'll leave it running for longer later.
Full size in spoiler:
With the Prime 95 the H100i temps maxed at 60 degrees which is good compared to the mid to high 70's the stock cooler did yesterday, that said it was only a 10 minute test so I'll leave it running for longer later.
The memory controller in ZEN+ (credits by The Stilt)
The memory controller in Pinnacle Ridge is identical to the one found in Raven. There are some differences in the software configuration, however the Phy IP itself is no doubt identical as the two share the same controller firmwares.
Compared to Summit Ridge, the revised controller in Pinnacle Ridge provides < 8.7% lower access latency on average (2133-3466MHz). The latency difference is largest at =< 2666MHz frequencies and starts to tail off at higher speeds.
Likewise, the SDF latency has slightly benefited from the changes. The average latency improvement (CCX2CCX latency) is < 2.2%, but just like with the memory latency the difference is tailing off as the memory speed increases. At 3200MHz MEMCLK the SDF latency difference almost falls within the margin of error.
Despite the extremely welcome latency improvement in Pinnacle Ridge, the memory latency is unfortunately still < 38% higher on average (2133-3466MHz) than on its closest rival from Intel (Coffee Lake).
While the changes to the memory controller in Pinnacle Ridge do provide lower latency, unfortunately the highest achievable memory frequency seems to be exactly the same as on Summit and Raven Ridge parts. A realistic expectation would be 3400 - 3533MHz depending on the silicon quality, the motherboard and the DRAM modules used. Some CPU specimens featuring an exceptional memory controller might be able to reach 3600MHz while maintaining true stability, however all of the tested 2700X samples were limited to 3400 - 3533MHz on both Crosshair VII Hero and MSI B350I PRO AC motherboards, regardless of the settings or the memory modules used. The stability was determined using “Ram Test” utility, which obviously sets the bar for stability a lot higher than the tests methods other reviewers typically use to deem the memory as “stable” at certain frequency.
On the tested samples, the distribution of the maximum achievable memory frequency was following:
3400MHz – 12.5% of the samples
3466MHz – 25.0% of the samples
3533MHz – 62.5% of the samples
There are clear differences in how the memory controller behaves on the different CPU specimens. The majority of the CPUs will do 3466MHz or higher at 1.050V SoC voltage, however the difference lies in how the different specimens react to the voltage. Some of the specimens seem scale with the increased SoC voltage, while the others simply refuse to scale at all or in some cases even illustrate negative scaling. All of the tested samples illustrated negative scaling (i.e. more errors or failures to train) when higher than 1.150V SoC was used. In all cases the maximum memory frequency was achieved at =< 1.100V SoC voltage.
AMD has revised the memory layout design guidance with Pinnacle Ridge targeting motherboards (i.e. 400-series) in an effort to potentially make the higher memory frequencies possible.
While this might theoretically improve the frequencies on some motherboards, generally the frequency limiting factor is the memory controller itself and not the topology the motherboard uses for memory signaling. Because of that, the newer 400-series motherboards alone should not be expected to provide improved memory frequencies at least by a significant margin.
The memory controller in Pinnacle Ridge is identical to the one found in Raven. There are some differences in the software configuration, however the Phy IP itself is no doubt identical as the two share the same controller firmwares.
Compared to Summit Ridge, the revised controller in Pinnacle Ridge provides < 8.7% lower access latency on average (2133-3466MHz). The latency difference is largest at =< 2666MHz frequencies and starts to tail off at higher speeds.
Likewise, the SDF latency has slightly benefited from the changes. The average latency improvement (CCX2CCX latency) is < 2.2%, but just like with the memory latency the difference is tailing off as the memory speed increases. At 3200MHz MEMCLK the SDF latency difference almost falls within the margin of error.
Despite the extremely welcome latency improvement in Pinnacle Ridge, the memory latency is unfortunately still < 38% higher on average (2133-3466MHz) than on its closest rival from Intel (Coffee Lake).
While the changes to the memory controller in Pinnacle Ridge do provide lower latency, unfortunately the highest achievable memory frequency seems to be exactly the same as on Summit and Raven Ridge parts. A realistic expectation would be 3400 - 3533MHz depending on the silicon quality, the motherboard and the DRAM modules used. Some CPU specimens featuring an exceptional memory controller might be able to reach 3600MHz while maintaining true stability, however all of the tested 2700X samples were limited to 3400 - 3533MHz on both Crosshair VII Hero and MSI B350I PRO AC motherboards, regardless of the settings or the memory modules used. The stability was determined using “Ram Test” utility, which obviously sets the bar for stability a lot higher than the tests methods other reviewers typically use to deem the memory as “stable” at certain frequency.
On the tested samples, the distribution of the maximum achievable memory frequency was following:
3400MHz – 12.5% of the samples
3466MHz – 25.0% of the samples
3533MHz – 62.5% of the samples
There are clear differences in how the memory controller behaves on the different CPU specimens. The majority of the CPUs will do 3466MHz or higher at 1.050V SoC voltage, however the difference lies in how the different specimens react to the voltage. Some of the specimens seem scale with the increased SoC voltage, while the others simply refuse to scale at all or in some cases even illustrate negative scaling. All of the tested samples illustrated negative scaling (i.e. more errors or failures to train) when higher than 1.150V SoC was used. In all cases the maximum memory frequency was achieved at =< 1.100V SoC voltage.
AMD has revised the memory layout design guidance with Pinnacle Ridge targeting motherboards (i.e. 400-series) in an effort to potentially make the higher memory frequencies possible.
While this might theoretically improve the frequencies on some motherboards, generally the frequency limiting factor is the memory controller itself and not the topology the motherboard uses for memory signaling. Because of that, the newer 400-series motherboards alone should not be expected to provide improved memory frequencies at least by a significant margin.
So basically Intel's IPC is only noticeably ahead in two tasks: CPU-bound gaming and Handbrake x265. Everything else is either a <4% win (e.g. Corona, Blender, single-threaded Cinebench), on par, or an small AMD win. Pretty much as expected considering the architectures, although I am not sure why x265 is still so bad on Ryzen.
Yes but that was a very forced demo. IPC matters when it matters. Intel can get to higher clockspeeds and AMD offers more cores for the same money.
It's an improvement that things are not clear cut anymore while still advancing.
Permabanned
- Joined
- 23 Apr 2014
- Posts
- 23,552
- Location
- Hertfordshire
"think" I am currently stable ish. Burn test passed and 200% hci memtest so far.
Most settings just left on Auto, PE level 3, -0.1 offset, 3333 memory. Stays well under 60 degrees benching.
Similar results to stock, though all core is over 4k, however think its a lot cooler.
May have a play with bclk later.
Most settings just left on Auto, PE level 3, -0.1 offset, 3333 memory. Stays well under 60 degrees benching.
Similar results to stock, though all core is over 4k, however think its a lot cooler.
May have a play with bclk later.
Of course, as I said earlier it's just an interesting exercise to see where the architectures currently sit and what their potential is. Based on what we've been told about 7 nm and what we've seen from Intel over the last 5 years, one would think AMD are more likely to catch up to Intel in terms of clock speed before Intel pulls away in terms of IPC. It also indicates that even if AMD does hit the 5 GHz mark with Ryzen 3, they still may lag Intel in gaming (although I doubt how much that'll matter for most gamers).
No he has 4000mhz ram at 4.5ghz or something XD the xmp makes 0 difference. B die is b die the chips are all the same and pcb makes basically no difference. I have hof 4000c19 and its not any faster than 3200c14 gskill on anything until you get to 4100c12 on Intel at 2v on the memory
Soldato
- Joined
- 2 Jan 2012
- Posts
- 12,255
- Location
- UK.
Not sure if poor attempt at trolling or genuinely don't know?
Anyway, memory frequencies are improved with Ryzen +. Seems to be limited return after 3200Mhz regardless. The jump from 2400Mhz to 3200Mhz is well worth it however.
Permabanned
- Joined
- 23 Apr 2014
- Posts
- 23,552
- Location
- Hertfordshire
I have yet to match my settings from the 1700X. However I got quite close.
So far.
1700X
2700X
So far.
1700X
2700X
I am 100 Percent NOT trolling. I am referring to comments I saw from AMD about memory frequencies being much improved and 3600 uowards was now much easier.
Also, just on my own expectations as well tbh.