- Joined
- Apr 26, 2008
- Messages
- 232 (0.04/day)
System Name | 3950X Workstation |
---|---|
Processor | AMD Ryzen 9 3950X |
Motherboard | ASUS Crosshair VIII Impact |
Cooling | Cryorig C1 with Noctua NF-A12x15 |
Memory | G.Skill F4-3600C16D-32GTZNC |
Video Card(s) | ASUS GTX 1650 LP OC |
Storage | 2 x Corsair MP510 1920GB M.2 SSD |
Case | Realan E-i7 |
Power Supply | G-Unique 400W |
Software | Win 10 Pro |
Benchmark Scores | https://smallformfactor.net/forum/threads/the-saga-of-the-little-gem-continues.12877/ |
I don't know how you some of you guys can claim "low IPC" by looking at this info.
IPC proxies which you can calculate from here are based on assumptions, not actual data.
Let me explain: Nobody knows at what "actual" clocks the Ryzen chip was running during the single-thread Cinebench run. Same goes for the Intel chips, among which it seems 6900K is the fitting contender for 1700X (8-core to 8-core).
Further, 1700X score of 154 was with a 2133MHz memory speed, while 6900K score of 153 was with 2400MHz (incidentally at same timings). Nobody knows if/how 1700X score would be impacted if it was run with a 2400Mhz memory.
Still, for the fun of it, let's do some calculations:
Anandtech disabled Turbo 3.0, so it is I suppose safe to assume that 6900K was running at its normal normal Turbo speed of 3.7GHz. So, per GHz, Broadwell did 153/3.7 = 41.3514
Even though we do not know, let's assume 1700X was working at its normal Turbo speed of 3.8GHz. So, per GHz, Ryzen did 154/3.8 = 40.5263
This means, relative to Broadwell, coupled with a slower memory, Ryzen single-thread CB15 performance at same clocks (again, not sure) is 98%.
Furthermore, do I have to remind everyone that this is a single-chip to single-chip comparison and with such small samples, 2% deviation could happen even if you used the same friggin chip on two different days???
From this limited data, anyone with a little sense would not find any reason to deny that Ryzen is equivalent to Broadwell in IPC performance.
Even when you compare the 1700X with the top 4-core Kaby Lake part (7700K, with 42.8889 per GHz score), you get 94.45% performance.
So you basically have an architecture that provides performance within 5% of the current top dog, hitting incredible price/performance and watt/performance levels, and you're not happy?
IPC proxies which you can calculate from here are based on assumptions, not actual data.
Let me explain: Nobody knows at what "actual" clocks the Ryzen chip was running during the single-thread Cinebench run. Same goes for the Intel chips, among which it seems 6900K is the fitting contender for 1700X (8-core to 8-core).
Further, 1700X score of 154 was with a 2133MHz memory speed, while 6900K score of 153 was with 2400MHz (incidentally at same timings). Nobody knows if/how 1700X score would be impacted if it was run with a 2400Mhz memory.
Still, for the fun of it, let's do some calculations:
Anandtech disabled Turbo 3.0, so it is I suppose safe to assume that 6900K was running at its normal normal Turbo speed of 3.7GHz. So, per GHz, Broadwell did 153/3.7 = 41.3514
Even though we do not know, let's assume 1700X was working at its normal Turbo speed of 3.8GHz. So, per GHz, Ryzen did 154/3.8 = 40.5263
This means, relative to Broadwell, coupled with a slower memory, Ryzen single-thread CB15 performance at same clocks (again, not sure) is 98%.
Furthermore, do I have to remind everyone that this is a single-chip to single-chip comparison and with such small samples, 2% deviation could happen even if you used the same friggin chip on two different days???
From this limited data, anyone with a little sense would not find any reason to deny that Ryzen is equivalent to Broadwell in IPC performance.
Even when you compare the 1700X with the top 4-core Kaby Lake part (7700K, with 42.8889 per GHz score), you get 94.45% performance.
So you basically have an architecture that provides performance within 5% of the current top dog, hitting incredible price/performance and watt/performance levels, and you're not happy?