Announcement

Collapse
No announcement yet.

Single Thread Score rating

Collapse
This is a sticky topic.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Single Thread Score rating

    Hi,
    What happen to the CPU Single Thread score rating? Why was it changed?
    CPU benchmarks are so useful for me when I spec systems for customers.. as I deal with all types of CPUs from Atom to Xeon.. having a quick look at the numbers (especially at single thread performance) gave me a good indication of the performance... I knew what to expect from which CPU without comparing it to anything else.. .and now.. :/

  • #2
    Hi,

    We released a new version of PerformanceTest a few days ago, version 10.
    Improvements in the benchmark test algorithms & using a more modern compiler resulted the single threaded test performing a much higher number of operations per second. These changes should push the CPU harder and use modern CPU features (out of order execution and multiple pipelines) better. The result was roughly 3x times more operations per second being performed, compared to PerformanceTest V9.

    Yesterday we started to switch over the graphs on the web site to start to use results from PerformanceTest V10. This accounts for the change in the results in the graphs.

    However in hindsight we think have done the wrong thing. We should had scaled down the PT10 single threaded result to match the PT9 results for the single threaded test. This single threaded test was already an average of values from several different single threaded algorithms. So additional scaling wouldn’t have changed the significance of the value.

    On Monday (9th March 2020) we plan to patch the version 10 release to scale the single threaded value back to the PT9 results. Things should then be back to normal. In the meantime we have reverted the single threaded graph on the web site to use only PT9 results.

    As we collect more PT10 results we expect PT10 to perform better on modern CPUs compared to older ones (relative to PT9). So overtime there might be a spreading out of the single threaded results, with the newer hardware pulling away from the older hardware a bit more.

    Sorry for any confusion all this has caused.

    More info:
    See also this post for some additional details
    https://www.passmark.com/forum/pc-ha...s-huge-changes

    Comment


    • #3
      Hi David,

      Thank you for the explanation. Good call with the results scaling. It makes so much easier to understand the numbers.
      Do you think that in the future it will be possible to for example filter the results to PT10 only?

      Regards,
      Les

      Comment


      • #4
        To be fair it really looks like you've been paid to tweak benchmark and boost Intel processors in the rating. Standings changed dramatically and now the Ryzen results are inconsistent and not very plausible. The PRO version of 3700 ryzen is 500 points ahead. How can you triple the algorithm performance using the new compiler? Have you used the intel c++ compiler known to use sub-optimal code for processors from vendors other than Intel? Have you triple checked the algorithm performance on different CPU and sure that there is no bug in it?

        Comment


        • #5
          Have you used the intel c++ compiler
          No. As per the PT10 release notes, Visual Studio 2019 is now used.
          We did try out the Intel compiler internally many years ago. Surprisingly it didn't actually favor Intel CPUs very much (much less than people who have never actually done the experiment expect). And it also didn't do greatly better than Visual Studio. We've been using Microsoft's compilers for every public release for the last 20 years. They are basically the standard for Windows development.

          How can you triple the algorithm performance using the new compiler?
          There were many changes. See the release notes.
          Variable alignment was a surprisingly big issue, as was rewriting bits of the code to allow more out of order instructions and allowing compiler optimization of code that was not previously optimized.

          Have you triple checked the algorithm performance on different CPU and sure that there is no bug in it?
          Yep, we are pretty sure it is OK. We;ve only got around 20 test machines in house. So testing top 100 CPUs isn't possible.
          Results for the top CPUs in single threaded is really really close. So it does't take much at all to move the ranking around. Just a few percent up or down. Expect the ranking to move around a bit more over the next week as new results come it before new average levels are found.

          But if it turns out the i9-9900KS is the new single threaded performance leader, we've got zero problems with that.

          Quotes:
          Hothardware.com
          "The Core i9-9900KS is the fastest processor we’ve tested for single and lightly-threaded workloads. It put up top marks in all of the single-threaded focused benchmarks"

          techradar.com
          "the best gaming chip"

          anandtech.com
          "In our benchmarks, we see it steaming ahead of any other consumer grade processor when it comes to single core performance"

          Comment


          • #6
            Originally posted by David (PassMark) View Post

            But if it turns out the i9-9900KS is the new single threaded performance leader
            9900 was the leader in the previous version of benchmark also. The question is why the new algorithm is so inconsistent on AMD CPUs. 3700 PRO is 20% faster than 3700X, the previous leader among AMD chips 3950X is now placed after the mobile Ryzen 3 4300U and lose to the most Intel chips as well. It looks like a weird bug, AVX is not used on some AMD chips or the single thread test is evaluated with some heavy second task running on the background.

            Comment


            • #7
              Some more concerns. Take a look on i9-9900KS vs Ryzen 7 3800X https://www.cpubenchmark.net/compare...00X/3593vs3499
              Both processors have the same number of cores. 9900KS is always boosted to higher frequency than 3800X.
              9900KS is 20% faster than 3800X in single thread benchmark but still lose in multithreaded benchmark!?
              How can it be? Math just don't sums up here. It should ether be slower in single thread test or should be much much faster running test on all cores.

              Comment


              • #8
                9900 was the leader in the previous version of benchmark also
                No it wasn't. Old V9 results can be found here.
                https://www.cpubenchmark.net/pt9_cpu_list.php
                It was very very close thing, but the 9900KS was actually 3rd. But the difference was so small it is meaningless.

                the previous leader among AMD chips 3950X
                No it wasn't. Previous leader was Ryzen 9 PRO 3900.

                The question is why the new algorithm is so inconsistent on AMD CPUs
                3700 PRO was fastest chip for V9. It is still the fastest AMD chip in V10, at least for the moment. But I suspect once we get in a few more V10 results for this rare CPU it's results will drop a few percent.

                In V9 the Ryzen 7 3800X was with 1.6% of the top score. In V10 Ryzen 7 3800X is within 16.2% of the top score. So yes, at least for the moment, there is a 14% relative drop for this AMD CPU in relative single threaded performance.

                Ryzen 3 4300U is super rare. We've got zero results for it in V10. Even in V9 we only had one. So this result isn't accurate (especially in V10).
                We'll look at doing some minor scaling on some of these rare CPUs.

                or the single thread test is evaluated with some heavy second task running on the background.
                No. It wouldn't be single threaded if there were multiple threads running.

                AVX is not used on some AMD chips
                The Single Threaded test is a combination of the Floating Point, Sorting and Compression tests.
                Floating Point & Sorting make zero use of AVX. Compression uses open source Gzip code. But I would be surprised if compiler would be making heavy use of SIMD / AVX. Pretty much impossible that it is using AVX512, which is the main set of instructions AMD is missing. So I don't think this is part of the reason.

                Comment


                • #9
                  Originally posted by David (PassMark) View Post

                  3700 PRO was fastest chip for V9. It is still the fastest AMD chip in V10, at least for the moment.
                  3900 PRO was second top, and now it is placed 31 losing 12% mark to 3700 PRO. 3950X was fourth fastest single threaded, now it is on 54 position.

                  3800X running 8 cores has cpu mark 23303. 9900KS running on faster base and turbo frequency only get 20400. At the same time 3800X is getting 20% lower points in single threaded test. How? This is weird. Single threaded test used to be an IPC performance benchmark and now it is like a random value +-20%. Well.. or it was random all these years.

                  Comment


                  • #10
                    Dear David,
                    Thanks you for your detailed reply, but I am not sure how is it possible that even the low ST boost clock Intel i3 9100F can have better ST score then the new Ryzen 3900X while we all know for a fact that Zen 2.0 has around ~8% better IPC then Intel current Coffee Lake, if you stand behind those scores then I must say that this new ST benchmark score must be flawed and not representative for real life performance gap between 4.2Ghz Coffee Lake vs ~4.5Ghz Zen 2.0.

                    I hope that it's just a bug and soon the scores will stabilized because there is no way that current Intel Coffee Lake is around ~20% better in ST then the new Zen 2.0.

                    Best Regards,
                    HwGeek.

                    Comment


                    • #11
                      So we've been doing some analysis overnight for the single threaded data. There is too much data to post it all. But here are a sub-set of the results to date for which we have a reasonable amount of confidence in (i.e. at least 3 samples to average).
                      CPU Model PT9 Single thread result Rank PT10 Single thread result Change in result
                      AMD Phenom II X4 965 1181.35 1371 1340.51 13%
                      Intel Core i7-8665U @ 1.90GHz 2314.40 147 2583.77 12%
                      AMD Athlon II X4 640 1024.19 1609 1141.63 11%
                      Intel Core i5-5200U @ 2.20GHz 1397.26 1107 1554.38 11%
                      Intel Core i5-760 @ 2.80GHz 1208.25 1336 1342.25 11%
                      Intel Core i3-380M @ 2.53GHz 970.04 1689 1075.29 11%
                      Intel Core i5-6300U @ 2.40GHz 1621.80 828 1795.17 11%
                      AMD Ryzen 5 3500U 1789.47 624 1976.34 10%
                      AMD Phenom II X4 955 1106.72 1487 1222.11 10%
                      Intel Core i5-6400 @ 2.70GHz 1824.15 588 2007.49 10%
                      Intel Core i7-6600U @ 2.60GHz 1802.63 611 1983.24 10%
                      Intel Core i5-7500T @ 2.70GHz 1855.82 560 2035.97 10%
                      Intel Core i7-6700 @ 3.40GHz 2154.55 248 2359.56 10%
                      Intel Core i7-5500U @ 2.40GHz 1545.55 929 1692.56 10%
                      Intel Core i7-6850K @ 3.60GHz 2191.78 214 2396.56 9%
                      Intel Core i5-5300U @ 2.30GHz 1533.89 946 1675.67 9%
                      Intel Core i7-6820HQ @ 2.70GHz 1873.25 534 2042.19 9%
                      Intel Core i7-860 @ 2.80GHz 1235.95 1305 1344.16 9%
                      Intel Core i5-6600K @ 3.50GHz 2145.68 255 2331.76 9%
                      Intel Core i5-560M @ 2.67GHz 1197.88 1353 1295.96 8%
                      Intel Core i7-7820HQ @ 2.90GHz 2048.52 348 2213.16 8%
                      Intel Core i3-5005U @ 2.00GHz 1091.94 1506 1179.06 8%
                      Intel Core i5-6600 @ 3.30GHz 2095.22 306 2262.03 8%
                      Intel Core i5-3337U @ 1.80GHz 1273.20 1257 1371.22 8%
                      Intel Core i3-6100U @ 2.30GHz 1300.80 1227 1400.28 8%
                      Intel Core i3-6006U @ 2.00GHz 1115.06 1464 1199.06 8%
                      Intel Core i5-6300HQ @ 2.30GHz 1672.59 765 1797.34 7%
                      Intel Core i7-640M @ 2.80GHz 1303.26 1223 1400.13 7%
                      Intel Core i7-6700K @ 4.00GHz 2354.04 128 2520.93 7%
                      Intel Core i5-6500 @ 3.20GHz 1948.45 447 2082.33 7%
                      Intel Core i3-8100 @ 3.60GHz 2097.86 301 2241.16 7%
                      Intel Core i5-7400 @ 3.00GHz 1944.59 455 2076.19 7%
                      AMD Ryzen 5 1600 1822.45 592 1945.76 7%
                      Intel Core i5-7500 @ 3.40GHz 2115.63 287 2258.53 7%
                      AMD Ryzen 7 3750H 1942.35 459 2072.64 7%
                      AMD Ryzen 5 3400G 2097.71 302 2238.38 7%
                      Intel Core i5-520M @ 2.40GHz 1061.55 1552 1130.27 6%
                      AMD Ryzen 3 1200 1719.31 710 1830.36 6%
                      AMD FX-6100 Six-Core 1187.51 1365 1262.08 6%
                      Intel Core i7-6700HQ @ 2.60GHz 1797.65 616 1909.61 6%
                      AMD Ryzen 5 2600X 2142.95 258 2274.71 6%
                      Intel Core i7-7700HQ @ 2.80GHz 1990.52 402 2110.86 6%
                      AMD Ryzen 5 2400G 1946.99 451 2064.53 6%
                      Intel Core i5-7300U @ 2.60GHz 1948.33 449 2064.71 6%
                      AMD Ryzen 3 3200G 1992.72 400 2109.53 6%
                      Intel Core i3-9100F @ 3.60GHz 2404.91 107 2544.95 6%
                      Intel Core i5-8265U @ 1.60GHz 2127.93 275 2249.29 6%
                      Intel Core i7-7700K @ 4.20GHz 2580.67 61 2726.71 6%
                      AMD Ryzen 3 2200G 1845.73 567 1946.68 5%
                      AMD Ryzen 5 1600X 1964.37 434 2070.49 5%
                      Intel Core i7-3610QM @ 2.30GHz 1637.90 815 1725.28 5%
                      Intel Core i9-7900X @ 3.30GHz 2447.39 96 2576.62 5%
                      Intel Core i7-7700 @ 3.60GHz 2342.44 135 2466.00 5%
                      Intel Core i7-4510U @ 2.00GHz 1628.77 825 1714.03 5%
                      Intel Core i7-3520M @ 2.90GHz 1783.73 630 1875.24 5%
                      AMD Ryzen 5 2500U 1668.35 770 1750.02 5%
                      Intel Core i5-4300U @ 1.90GHz 1562.86 907 1639.30 5%
                      AMD Ryzen 7 2700X 2182.17 219 2286.22 5%
                      Intel Core i5-1035G4 @ 1.10GHz 2406.59 106 2518.03 5%
                      Intel Core i5-4200U @ 1.60GHz 1370.08 1147 1431.74 5%
                      Intel Core i3-550 @ 3.20GHz 1299.71 1228 1355.69 4%
                      AMD Ryzen 5 2600 2004.09 386 2089.57 4%
                      Intel Core i7-9750HF @ 2.60GHz 2467.68 88 2571.24 4%
                      Intel Core i5-9500 @ 3.00GHz 2546.64 66 2652.32 4%
                      Intel Pentium Gold G5400 @ 3.70GHz 2172.79 229 2260.79 4%
                      AMD Ryzen 7 1700X 1886.50 524 1962.49 4%
                      Intel Core i7-5600U @ 2.60GHz 1662.18 774 1727.80 4%
                      Intel Core i3-540 @ 3.07GHz 1223.15 1317 1270.81 4%
                      Intel Core i5-9600K @ 3.70GHz 2678.09 39 2782.13 4%
                      Intel Core i7-930 @ 2.80GHz 1219.50 1322 1265.09 4%
                      Intel Core i7-8750H @ 2.20GHz 2340.88 136 2427.69 4%
                      Intel Core i5-3230M @ 2.60GHz 1576.45 889 1633.75 4%
                      Intel Core i3-4030U @ 1.90GHz 1042.44 1581 1079.90 4%
                      Intel Core i9-9900 @ 3.10GHz 2794.83 20 2894.59 4%
                      Intel Core i3-7100 @ 3.90GHz 2225.94 195 2305.24 4%
                      AMD Ryzen 7 2700 1982.52 413 2051.47 3%
                      Intel Core i7-950 @ 3.07GHz 1327.62 1194 1373.21 3%
                      Intel Core i7-9850H @ 2.60GHz 2520.77 74 2607.13 3%
                      Intel Core i9-9900KF @ 3.60GHz 2884.94 15 2982.94 3%
                      AMD Ryzen 7 1800X 1970.36 428 2036.96 3%
                      AMD Ryzen Threadripper 1950X 1989.92 403 2056.77 3%
                      Intel Core i5-8259U @ 2.30GHz 2211.32 200 2284.83 3%
                      Intel Core i7-8565U @ 1.80GHz 2298.25 152 2370.67 3%
                      Intel Core i7-9750H @ 2.60GHz 2464.16 91 2541.00 3%
                      Intel Xeon W-2123 @ 3.60GHz 2144.54 256 2211.07 3%
                      Intel Core i7-3630QM @ 2.40GHz 1685.18 750 1736.71 3%
                      AMD FX-8350 Eight-Core 1510.00 976 1555.06 3%
                      Intel Core i5-8600K @ 3.60GHz 2511.37 78 2583.85 3%
                      Intel Core i5-3317U @ 1.70GHz 1215.03 1327 1249.35 3%
                      Intel Core i7-9700 @ 3.00GHz 2690.32 36 2765.66 3%
                      AMD Ryzen 5 PRO 3500U 1895.68 513 1947.82 3%
                      AMD Ryzen Threadripper 2950X 2207.52 204 2267.97 3%
                      Intel Core i7-3820QM @ 2.70GHz 1844.09 569 1892.53 3%
                      Intel Core i5-4210U @ 1.70GHz 1430.46 1066 1467.60 3%
                      Intel Core i7-4720HQ @ 2.60GHz 1909.52 496 1955.94 2%
                      AMD FX-8370 Eight-Core 1537.08 942 1574.40 2%
                      Intel Core i7-9700KF @ 3.60GHz 2793.28 21 2860.89 2%
                      Intel Core i7-8700 @ 3.20GHz 2624.50 50 2686.85 2%
                      Intel Core i7-6500U @ 2.50GHz 1640.39 813 1679.28 2%
                      Intel Core i9-9900K @ 3.60GHz 2888.73 14 2954.55 2%
                      Intel Core i7-8700K @ 3.70GHz 2696.43 34 2757.30 2%
                      Intel Pentium 5405U @ 2.30GHz 1331.05 1192 1360.39 2%
                      Intel Core i5-9600KF @ 3.70GHz 2659.62 42 2717.87 2%
                      Intel Core i5-9400F @ 2.90GHz 2392.99 112 2444.80 2%
                      Intel Core i5-10210U @ 1.60GHz 2300.07 150 2348.15 2%
                      Intel Core i5-9300H @ 2.40GHz 2348.32 131 2395.21 2%
                      Intel Core i7-9700K @ 3.60GHz 2812.37 18 2867.92 2%
                      AMD FX-9590 Eight-Core 1724.58 701 1758.15 2%
                      AMD FX-6300 Six-Core 1406.96 1089 1434.16 2%
                      Intel Core i9-9900KS @ 4.00GHz 2992.34 3 3048.18 2%
                      Intel Core i5-7300HQ @ 2.50GHz 1856.87 558 1888.02 2%
                      Intel Core i5-750 @ 2.67GHz 1145.91 1414 1164.15 2%
                      AMD Ryzen 7 PRO 3700U 1902.95 501 1931.20 1%
                      Intel Core i3-8130U @ 2.20GHz 1917.02 488 1944.24 1%
                      Intel Core i7-8650U @ 1.90GHz 2208.18 202 2238.56 1%
                      Intel Core i7-10510U @ 1.80GHz 2430.48 103 2460.29 1%
                      Intel Xeon E5450 @ 3.00GHz 1265.42 1271 1279.92 1%
                      AMD Ryzen 5 3550H 1872.30 536 1892.85 1%
                      Intel Core i7-10710U @ 1.10GHz 2384.66 116 2410.09 1%
                      Intel Core i5-8250U @ 1.60GHz 1923.42 480 1940.38 1%
                      Intel Xeon E3-1230 V2 @ 3.30GHz 1950.26 445 1964.10 1%
                      Intel Xeon E-2274G @ 4.00GHz 2727.63 30 2744.80 1%
                      Intel Core i5-8400T @ 1.70GHz 1921.59 484 1929.41 0%
                      Intel Core i5-3570 @ 3.40GHz 2022.68 370 2023.00 0%
                      Intel Xeon E3-1240 V2 @ 3.40GHz 2027.48 365 2023.83 0%
                      Intel Core i5-8500 @ 3.00GHz 2404.57 108 2399.51 0%
                      Intel Xeon E5-2667 v2 @ 3.30GHz 2010.74 378 2006.42 0%
                      Intel Core i5-3330 @ 3.00GHz 1700.04 732 1695.70 0%
                      Intel Core i5-3320M @ 2.60GHz 1642.32 811 1637.91 0%
                      Intel Core2 Quad Q9400 @ 2.66GHz 1127.68 1439 1121.53 -1%
                      Intel Core i7-3770S @ 3.10GHz 2025.78 367 2012.00 -1%
                      Intel Core i5-3360M @ 2.80GHz 1763.13 652 1746.45 -1%
                      Intel Core i5-8400 @ 2.80GHz 2330.63 143 2305.94 -1%
                      Intel Core i7-1065G7 @ 1.30GHz 2518.80 75 2486.49 -1%
                      Intel Core i5-3470 @ 3.20GHz 1916.03 490 1887.81 -1%
                      Intel Core2 Duo P8700 @ 2.53GHz 993.35 1652 978.12 -2%
                      Intel Core i5-1035G1 @ 1.00GHz 2329.60 144 2292.83 -2%
                      Intel Core i5-8300H @ 2.30GHz 2294.68 153 2254.58 -2%
                      Intel Core i3-4150 @ 3.50GHz 2009.46 380 1974.18 -2%
                      Intel Core i5-4690K @ 3.50GHz 2235.34 186 2195.31 -2%
                      AMD Ryzen 7 1700 1781.15 634 1747.56 -2%
                      Intel Core i7-5930K @ 3.50GHz 2103.31 293 2059.42 -2%
                      AMD FX-6350 Six-Core 1484.41 1013 1453.05 -2%
                      Intel Core i7-2670QM @ 2.20GHz 1394.20 1112 1362.97 -2%
                      Intel Core i3-4130 @ 3.40GHz 1958.37 441 1911.68 -2%
                      Intel Core i5-4670K @ 3.40GHz 2186.87 217 2133.26 -2%
                      Intel Core i5-4570 @ 3.20GHz 2057.24 343 2004.23 -3%
                      Intel Core i5-3570K @ 3.40GHz 2030.05 362 1975.45 -3%
                      Intel Core i5-6200U @ 2.30GHz 1496.20 1001 1453.65 -3%
                      Intel Core i3-3240 @ 3.40GHz 1802.72 610 1751.47 -3%
                      Intel Core i7-3770 @ 3.40GHz 2068.11 335 2008.14 -3%
                      Intel Core i5-4460 @ 3.20GHz 1942.49 457 1884.68 -3%
                      Intel Core i7-3770K @ 3.50GHz 2083.89 317 2017.98 -3%
                      Intel Core i3-3220 @ 3.30GHz 1755.62 655 1695.15 -3%
                      Intel Core i5-2410M @ 2.30GHz 1325.72 1195 1279.99 -3%
                      Intel Core i5-9400 @ 2.90GHz 2390.66 115 2307.29 -3%
                      AMD A9-9425 1466.85 1027 1414.39 -4%
                      Intel Xeon E3-1230 v3 @ 3.30GHz 2094.96 307 2019.30 -4%
                      Intel Core i7-4770 @ 3.40GHz 2227.54 193 2139.53 -4%
                      Intel Core i5-4590 @ 3.30GHz 2117.21 284 2028.58 -4%
                      Intel Core i5-4690 @ 3.50GHz 2218.49 198 2123.53 -4%
                      Intel Core i7-4700MQ @ 2.40GHz 1809.02 606 1712.12 -5%
                      Intel Core i5-7600K @ 3.80GHz 2380.52 117 2251.74 -5%
                      AMD A8-7600 APU 1542.56 937 1458.70 -5%
                      Intel Core i5-7200U @ 2.50GHz 1714.41 716 1619.29 -6%
                      Intel Core i7-4790K @ 4.00GHz 2528.45 71 2382.61 -6%
                      Intel Core i7-4800MQ @ 2.70GHz 1992.73 399 1872.97 -6%
                      Intel Pentium B960 @ 2.20GHz 1069.74 1533 1002.51 -6%
                      Intel Pentium G3240 @ 3.10GHz 1794.93 618 1679.56 -6%
                      Intel Celeron N3350 @ 1.10GHz 766.43 2024 709.58 -7%
                      Intel Core i7-4790 @ 3.60GHz 2282.67 159 2086.23 -9%
                      Intel Core i3-2100 @ 3.10GHz 1584.45 876 1447.59 -9%
                      Intel Core i9-9880H @ 2.30GHz 2530.11 68 2310.25 -9%
                      Intel Core i7-4600U @ 2.10GHz 1690.38 745 1542.78 -9%
                      AMD A10-8700P 1220.24 1320 1111.05 -9%
                      Intel Core i7-3740QM @ 2.70GHz 1865.11 545 1695.01 -9%
                      Intel Core i3-7020U @ 2.30GHz 1288.96 1241 1169.76 -9%
                      Intel Core i3-2120 @ 3.30GHz 1690.58 744 1528.74 -10%
                      Intel Core i3-2310M @ 2.10GHz 1008.12 1629 906.01 -10%
                      Intel Core i3-4160 @ 3.60GHz 2070.75 333 1859.31 -10%
                      Intel Xeon E5-2643 @ 3.30GHz 1812.92 603 1625.12 -10%
                      Intel Xeon E5-1650 v2 @ 3.50GHz 2003.99 387 1784.50 -11%
                      AMD FX-8320 Eight-Core 1395.50 1110 1241.97 -11%
                      Intel Core i5-3210M @ 2.50GHz 1520.84 963 1349.94 -11%
                      Intel Core i5-2400 @ 3.10GHz 1747.92 675 1550.12 -11%
                      Intel Core i5-2500K @ 3.30GHz 1899.42 506 1680.89 -12%
                      Intel Core i7-3930K @ 3.20GHz 1935.10 469 1710.35 -12%
                      Intel Core i5-2520M @ 2.50GHz 1492.05 1006 1318.68 -12%
                      Intel Core i7-5820K @ 3.30GHz 2025.12 368 1788.71 -12%
                      Intel Xeon E5-2689 @ 2.60GHz 1753.23 660 1547.06 -12%
                      Intel Core i7-2600K @ 3.40GHz 1940.67 463 1702.64 -12%
                      Intel Xeon E5-1620 @ 3.60GHz 1948.33 448 1707.82 -12%
                      AMD Ryzen Threadripper 3960X 2940.02 7 2572.10 -13%
                      Intel Core i7-960 @ 3.20GHz 1389.73 1121 1215.64 -13%
                      Intel Core i7-2600 @ 3.40GHz 1922.01 483 1679.00 -13%
                      Intel Core i7-4770K @ 3.50GHz 2249.80 176 1961.85 -13%
                      AMD Ryzen 9 3950X 2980.43 4 2585.38 -13%
                      Intel Core i7-8550U @ 1.80GHz 2122.72 279 1840.38 -13%
                      Intel Core i7-7500U @ 2.70GHz 1910.50 495 1650.77 -14%
                      AMD Ryzen 9 3900X 2938.84 8 2532.66 -14%
                      Intel Core i3-4170 @ 3.70GHz 2128.08 273 1833.47 -14%
                      Intel Xeon E5-1650 @ 3.20GHz 1940.96 462 1668.80 -14%
                      AMD Ryzen 7 3700X 2905.79 13 2480.55 -15%
                      AMD Ryzen 9 3900 2952.38 6 2515.29 -15%
                      Intel Core i3-4005U @ 1.70GHz 939.31 1733 793.79 -15%
                      AMD Ryzen 5 3600 2803.98 19 2368.69 -16%
                      Intel Core i7-4810MQ @ 2.80GHz 2072.55 329 1749.48 -16%
                      AMD Ryzen 5 3600X 2911.51 11 2457.37 -16%
                      AMD Ryzen Threadripper 3970X 2936.48 9 2476.57 -16%
                      AMD Ryzen 5 3500X 2770.94 23 2328.78 -16%
                      Intel Core i7-8850H @ 2.60GHz 2432.64 102 2035.13 -16%
                      AMD Ryzen 7 3800X 2978.64 5 2490.83 -16%
                      So new algorithms benefit some CPU families and hurt other CPU families. Overall the the average movement is close to 0% however (the gains equal the losses).

                      Looking just at the Ryzen results, as these seen to have been the most controversial.

                      The Ryzen 2000 series did better, while the 3000 series did worse. (You might be thinking that the AMD Ryzen 5 3500U & 3400G breaks this rule, but it seems AMD released their 3000 series mobile parts with the 3xxx name, even thought they were based on 2xxx series technology). So these are really 2000 series parts as well.

                      So comparing a 2000 series to a equivalent 3000 series in PT9 and PT10 we see.

                      AMD Ryzen 5 2600
                      PT9 single threaded score: 2004.09
                      PT10 single threaded score: 2089.57

                      AMD Ryzen 5 3600
                      PT9 single threaded score: 2803.98 (+40% generation gain)
                      PT10 single threaded score: 2368.69 (+13% generation gain)

                      So from the results above it seems in V9 we claimed new generation 3000 series was 40% faster for single threaded. Clearly this wasn't very reflective of real life. The 3000 series wasn't 40% faster, except maybe in a few edge cases. Now with V10 results we are claiming the 3000 series is 13% faster than the 2000 series. Which seems a much more plausible number and more reflective of real life applications. So I think from that point of view the new results are an definite improvement (although Ryzen 3000 owners might be slightly miffed. 2000 series owners probably won't complain however).

                      Comment


                      • #12
                        Next there is the question of should both the 2000 and 3000 series be higher up the new PT10 single thread rankings (i.e. above the best Intel CPUs)?
                        There is no right answer for this as performance really depends on the particular application you are running. The best Intel and AMD CPUs are close enough at the moment that you can make a case either to be faster based on the applications you choose to run.

                        Looking at the current results today we have.
                        (These will change slightly over the coming weeks however)

                        Intel Core i7-9700K @ 3.60GHz
                        PT10 Single threaded result: 2812

                        AMD Ryzen 7 2700X
                        PT10 Single threaded result: 2182 (22% slower)

                        AMD Ryzen 5 2600X
                        PT10 Single threaded result: 2142 (23% slower).

                        How does this compare to real life? Here is a result from Tom's Hardware running the POV-RAY application, single threaded.

                        Click image for larger version

Name:	SingleThreadPT10-1.png
Views:	9547
Size:	199.8 KB
ID:	46875

                        So for this real life app we see the following
                        (remember lower score are better for this POV rendering test)

                        Intel Core i7-9700K @ 3.60GHz
                        POV-Ray result: 505

                        AMD Ryzen 7 2700X
                        POV-Ray result: 633 (25% slower)

                        AMD Ryzen 5 2600X
                        POV-Ray result: 636 (26% slower)

                        So these results line up pretty well with PT10.

                        Cinebench (single threaded) gives results of 16% and 18%.
                        Y-Cruncher (single threaded) gives results of 88% and 95% (big difference is due to AVX instructions in Intel's CPU)

                        So if we accept that the Ryzen 3000 series gives +13% performance over the 2000 series, then that still puts them around 10 to 15% behind the best Intel CPUs.

                        I am sure it is possible cheery pick counter examples, but hopefully the majority of people will see the new results as an improvement over what we had.

                        This doesn't totally address everything however. We still have a problem presenting results for which we have zero PT10 samples (the rare and old CPUs). We are looking at that problem at the moment. Hopefully we'll have a reasonable solution in the next couple of days and get those rare units into their correct rankings.




                        Comment


                        • #13
                          Dear David,
                          As you see the problem Zen 2.0 scores, even if you compare the bench from TH, the Zen 2.0 ST performance is better then Coffee Lake and only the 5Ghz 9900K/S can match it.
                          on AVG the 9900K and Ryzen 3900X/3950X should be in ~5% margin, not ~20% like the PT10 ST scores show.
                          Here same Pov-Ray benchmark:

                          Best Regards,
                          HwGeek.
                          Click image for larger version

Name:	UV65xPCzgrKpK3QFDnz2YK-650-80.png
Views:	8730
Size:	67.3 KB
ID:	46883
                          Click image for larger version

Name:	vBFBd9MiDQeqgfZtAGrPPK-650-80.png
Views:	8479
Size:	68.7 KB
ID:	46884
                          Click image for larger version

Name:	szoccWxuqz2yKfPbfYjXJS-650-80.png
Views:	8535
Size:	67.2 KB
ID:	46885

                          Comment


                          • #14
                            Perhaps a difference in L3 cache size could explain why the Zen 2 cpus are performing so poorly compared to their normal performance? The 4300u is far ahead of their desktop counterparts which makes no sense given the reduction in clock speed, and the main differentiator is a reduction in L3 cache per core(4x reduction). I don't know what would cause that in your suite but you might be able to figure it out.

                            Comment


                            • #15
                              Ok. Now with more sample data there is no weird spikes among AMD CPU. The fact that ThreadReapers perform better than the boosted to bigger frequency 3950X... one can claim that this is due to better thermal design of threadreapers and the test is slow and very tough. etc etc.

                              All your explanation of how Zen 2 is worse than Intel just can't explain why they are still better multithreaded. 9900KS just can't perform worse than 3800X then. It is boosted to 5 GHz all cores and has the same number of cores. If it is 20% faster in ST it should just rip 3800X to shreds multithreaded. So why it lose to 3800X in multithreaded test then?

                              Comment

                              Working...
                              X