Hm then I think the surprising difference for one core is caused by the type of audio interface: USB still has serious amount of cpu overhead, while I use firewire (on macos), which has very little overhead. I set my cpu timing to “turbo” in the bios, I don’t know if that means overclocked, but it shows the cpu type typical frequencies in the OS.
I think next time I really will buy such an AMD cpu. Also supported by hackintosh btw.
Hehe, who knows what they come up with next at AMD…And really, it is not surprising at all, as I´ve mentioned, the single core speed of the TR 1950X is equal to the I7 4XXX CPUs, just that you get 16 cores + 16 HT threads on the TR 1950X compared to “just” 4 or 6 (8 / 12) on the I7 CPUs 
Also, there are differences in the compiler/efficiency how certain commands are translated for the CPU (and AMD Threadripper is not fully supported by the compilers yet, so as soon as this happens, speeds will get even better too). But you can hunt me with anything *tosh, I am all Windows and hate MAC´s + Apple, haha
Win 10 64 bit - Renoise 3.1 - AMD RYZEN 5 1600 STOCK
ReBench - Version extended to 256 tracks and 1537 effects
10ms latency
1 CORE - 47 RPTS
6 CORES - 214 RPTS
100ms latency
1 CORE - 73 RPTS
6 CORES - Ta! Amiga Rulez!!! 59% CPU usage
Ryzen 2600 at stock, 16GB RAM at 2993Mhz
Win 10 64bit - Renoise 3.1
ReBench 256
Direct Sound:
10ms:
Single Core - 48 RPTS
6 Core - 194 RPTS
100ms:
Single Core - 77 RPTS
6 Core - Ta! Amiga Rules!!! (Max CPU 56%)
Babyface Pro ASIO
256 samples (5.8ms):
Single Core - 57 RPTS
6 Core - 256 RPTS (hit 90% on last pattern!)
512 samples (11.6ms):
Single Core - 71 RPTS
6 Core - Ta! Amiga Rulez!!! (Max CPU 73%)
Thanks to Cactoos for the great benchmark!
AMD Threadripper 1950x @ 3.8Ghz
ReBench.xrns - Passed all patterns with max load 37%
ReBench_256.xrns - Passed all patterns with max load 56%.
Rebenchmarked my system with Renoise 3.4 using the rebench_x256.xrns.
i7 4770, macos 10.14.6 at 44.1khz, 8ms, 4 core (8 virtual), 4k->4.5k resolution upscaling(heavy cpu load), 90% overload protection setting:
Renoise 3.4: 248 RPTS (it started to crackle around 190)
Renoise 3.3: 153 RPTS (crackling started around 140)
Renoise 3.1 with 1080p monitor, 10ms, osx 10.9.4: 207 RPTS
M1 Max (MBP 14) - passed (max cpu usage ~61%) all patterns from ReBench_x256 using built-in audio, 2ms latency and with dithering on, on Renoise 3.4 ARM with Metal rendering.
Macbook 2020 M1 “normal” lol
renoise 3.4
dither off 90% warning
rane 72 soundcard
rebench 256
some limited amount of backround processes, about 10% maybe 20% of one cpu, normal usage situation
latency 01 8 cpu core = 217
latency 10 8 cpu core = PASS ALL 256 max 70%
latency 10 1 cpu core= 95
And real live comparision vs 2012 mid macbook. Orchestral project that i needed to render parts and use renders, freeze etc, in native mode with only few bridged plugins is around 40%cpu and a lot lower latency.
Rebench 256
Intel(R) Xeon(R) CPU E3-1270 v5 @ 3.60GHz (win10 pro with 64 GB ram) - bought 2017… Dell…
Renoise 3.3
Latency. 10ms
1cpu - trk 41
8cpu - trk 144
Latency: 100ms
1cpu - trk 67
8cpu - trk ALL (began playing from 1 again)
Renoise 3.4
Latency: 10ms
1cpu - trk 36
8cpu - trk 180
Latency: 100ms
1cpu - trk 68
8cpu - trk 249
Since the 3.3 test was made after i did the 3.4 test i re-ran the 3.4 test in which it passed with 8 cpu and 100ms… Which has lead me to belive it might have been an unlycky coincidence where the scan of midi-devices occured at about track 249 the last time…
Interesting would be: benchmark in 3.3 and then 3.4.
re-ran mine…
AMD Ryzen 9 3900X 12-Core Processor, 3800 Mhz, Win 10 Pro, 64 GB RAM, 44100 kHz, Focusrite ASIO.
Multicore:
Latency: 1.5 ms
All pass: 55% CPU.
Latency: 5,8 ms
All pass: 41% CPU.
Single core:
Latency: 5.8 ms
85 RPTS
3.3.2 and 3.4 perform exactly the same.
Intel i5-9600k, 3.7GHz, 6 Cores, Win 10 Pro, 16 GB RAM, Renoise 3.4
1 CPU: 63
6 CPU: 227
My CPU is almost perfect for my music but I could use one or two more cores to be fully comfortable
ReBench_256.xrns
CPU: _____________ Intel i9-12900K | 8x P-Core 4.9-5.2Ghz + 8x E-Core 4.0Ghz | 24 Threads
OS: ______________ Win 11
Interface: _______ ASIO
Latency: _________ 5 ms
Renoise 3.3, 01 CPU: 161 RPTS
Renoise 3.3, 02 CPU: All Pass w/ 82.2% peak CPU Load
Renoise 3.3, 24 CPU: All Pass w/ 26.0% peak CPU Load
Renoise 3.4: same as 3.3
CPU: _____________ Intel i7-5820K | 6x 4.3Ghz | 12 Threads
OS: ______________ Win 11
Interface: _______ ASIO
Latency: _________ 5 ms
Renoise 3.4, 01 CPU: 68 RPTS
Renoise 3.4, 02 CPU: 128 RPTS
Renoise 3.4, 12 CPU: All Pass w/ 74.0% peak CPU Load
CPU: _____________ AMD Ryzen 3600 | 6x 3.9-4.2Ghz | 12 Threads
OS: ______________ Win 10
Interface: _______ ASIO
Latency: _________ 5 ms
Renoise 3.4, 01 CPU: 78 RPTS
Renoise 3.4, 02 CPU: 140 RPTS
Renoise 3.4, 12 CPU: All Pass w/ 75.0% peak CPU Load
CPU: _____________ Intel i5-4670K | 4x 3.6-3.8 Ghz | 4 Threads
OS: ______________ Win 10
Interface: _______ ASIO
Latency: _________ 5 ms
Renoise 3.4, 01 CPU: 52 RPTS
Renoise 3.4, 02 CPU: 78 RPTS
Renoise 3.4, 04 CPU: 163 RPTS
Renoise 3.4.1 (multicore = on / pdc = on)
142
Audio UMC Asio running at 48 kHz.
Buffersize: 128 samples (~2,7 ms).
OS: Win 10 Pro 21H2 / .1620
Board: MS-7B48 (MSI Z370-A Pro)
BIOS-Version/-Datum American Megatrends Inc. 2.C0, 04.06.2020
SMBIOS-Version 2.8
Processor: Intel(R) Core™ i7-9700K CPU @ 3.60GHz, 3600 MHz, 8 Kern(e), 8 logische(r) Prozessor(en)
RAM: Installierter physischer Speicher (RAM) 48,0 GB (DDR4) 1,2 gHz Dram / Uncore ~4,3 gHz
Pagefile: Größe der Auslagerungsdatei 14,0 GB
Gfx: Intel UHD 630 / Nvidia RTX2080 (8 GB)
The test stopped at 68 % due to activated overload prevention in the preferences.
Wow, that’s pretty impressive. Especially the single core performance.
The sound of this song reminds me of an old paper press machine for printing 
that’s exactly why i wanted to get one in the first place.
this CPU is totally bonkers in terms of single-threaded performance, especially since you can easily let it boost to 5.5-5.7Ghz for single core usage and take back the clock and voltage for heavy load (4+ threads).
perfect for renoise and realtime audio in general.
I was interested to compare these to the M1 Pro:
Rebench 256, M1Pro, macos 10.16.1, Renoise 3.4.1, 5 ms, 44.1kHz:
1 core: 97 RPTS
2 cores: 177 RTPS
3 cores: 256 RTPS
10: cores: all passed with peak of 55%
Indeed that Intel CPU has an impressive single core performance. But keep in mind that M1Pro is a mobile cpu, and the fan did not even seem to have kicked in. Maybe this is a sign that Renoise still could use the M1Pro even more efficiently. This performance is more than enough for me. I understand though that Apple is not attractive for a lot of people and always comes with annoying compromises. I wonder how your CPU would perform under macos/hackintosh, since after reading the numbers here, it seems to me that the macos Renoise actually can perform a bit better than the Windows version…?
Single core result would be interesting, too! Maybe with 5ms / 44.1kHz, just like above, so we could more accurately compare here.