Earlier this month Intel officially announced their next generation of desktop processors. They were formally known as Arrow Lake and now have their new CPU naming as their Core Ultra 200 Series of CPUs. These are the successors to the 14^th-generation Core CPUs and the new naming is going to be a little confusing to some as it is not only the addition of Ultra in the naming but the CPU numbers themselves which have followed the same format for 14 generations are changing as well. This includes dropping the number of numbers in the CPU name down to three from 5. The two CPUs that they sent over for testing at launch are the Core Ultra 9 285K and the Core Ultra 5 245K. This generation's focus according to Intel is to reset their performance per watt trajectory. They have a new memory controller, new P and E core microarchitectures, larger P-core L2 cache, and more. So let’s dive in and see how the new CPUs perform!

Product Name: Intel Core Ultra 200S

Review Samples Provided by: Intel

Written by: Wes Compton

Amazon Affiliate Links:

Intel Core Ultra 9 285K

Intel Core Ultra 5 245K

 

Intel Core Ultra 200S

All of the information on the new CPUs came out back on the 10^th, so a majority of you already know all about the new Core Ultra 200 Series of CPUs. But for those of you who don’t, I did want to touch on a few things and also talk about the Core Ultra 9 285K and Core Ultra 5 245K that I will be testing today. With the Arrow Lake architecture, intel’s program goals were to improve their performance per watt and to also improve the overall user experience which as they have listed in their slide below part of that is dropping package temperatures when gaming so both of those tie in together.

Arrow Lake consists of Intel’s Skymont e-cores and Lion Core P-Cores. In their breakdown, they touch on the new e-cores having twice the L2 cache bandwidth and having a 32% IPC uplift over the last generation. The new P-Cores don’t have as significant of an uplift but still see 9%.

While I am only testing two, Intel is launching five CPUs in total. There is the Ultra 9 285K at the top with 24 cores (8 performance and 16 e-cores) and a max clock speed of 5.7GHz, it has a suggested retail price of $589. The next step down is the Ultra 7 265K with 20 cores (8 performance and 12 e-cores) and a max clock speed of 5.5 GHz for $394. Below that they have the same CPU but in the KF model meaning it doesn’t have onboard video for $379. Then we have the Ultra 5 245K with 14 cores (6 p-core and 8e-cores) with a max clock speed of 5.2 GHz for $309. Then last up is a second 245 but the KF model again without the iGPU for $294. It’s interesting to see how the 265K and 285K both have the same number of p-cores but just have the slight clock speed change and the four additional e-cores. Another big note that some might not notice is that hyper-threading is no longer there. The total thread count on all of these matches the core count. Beyond core count and clock speed, they did ramp up the L2 and Smart cache depending on the number of cores. The max turbo wattage changes depending on the CPU as well with the 285K and 265K having 250 watts and the 245K lower at 159 watts. For the models with Xe Cores for the iGPU, they all have the same 4 cores available but the 245K does have a barely lower clock speed at 1.9 GHz vs 2 GHz on the other two.

Of course, I also wanted to show off Intel’s press kit as they have clearly put a lot of time into it. The two CPUs came in a large box decked out with Intel branding. Along with the two CPUs they also included an acrylic brick with an information card showing off details on the new CPUs.

With a new socket, the new CPUs have a new pinout and different notches to prevent it from being put in the wrong motherboard but they are the same dimensions. Below I have a comparison between the two.

For testing, Corsair sent over one of their new 6400MT/s Pro Series kits. I love how simple and clean the heat spreaders on these look. They are blacked out other than the Crucial logo on one side and on the back side they say Crucial DDR5 Pro and have a small sticker with the required certification logos, model information, and serial numbers on them.

 

Test Rig and Procedures

Test System

Motherboard: GIGABYTE Z890 AORUS PRO ICE- Live Pricing

Cooling: Enermax LIQMAXFLO 360mm Liquid CPU Cooler - Live Pricing

Noctua NT-H1 Thermal Paste - Live Pricing

Memory: Crucial Pro OC 32GB 6400 MHz CP2K16G64C38U5B - Live Pricing

Storage: Crucial T700 2TB - Live Pricing

Video Card: Nvidia RTX 4090 FE - Live Pricing

Power Supply: Corsair AX1200 - Live Pricing

Case: Primochill Wetbench - Live Pricing

OS: Windows 11 Pro 64-bit - Live Pricing

 

CPU Testing Procedures
Blender	Using the official benchmark using 3.6
Handbrake	Tested using 2d 4k 60 FPS version, resized using the Fast 1080p30 preset, average FPS used. Here is the file download location http://bbb3d.renderfarming.net/download.html
Cinebench R20	CPU and CPU (Single Core results)
Cinebench R23	CPU and CPU (Single Core results)
Cinebench 2024	Multi and Single Core tests
POV-Ray	Using built-in benchmark. Settings are all set to the "Standard Benchmark Switches" listed on http://www.povray.org/download/benchmark.php
V-Ray Benchmark 1.0.8	CPU Benchmark time used as the score
V-Ray Benchmark 5.0.2	vsamples score
CPUz	CPUz's built-in benchmark on 17.01.64 version, both single and multi-core scores
wPrime	1024M Setting, core count set to the number of threads the CPU has
7-Zip	Built-in benchmark set to 32MB, which runs 10 passes. Combined MIPS, Compressing, and Decompressing scores were all used. Just the Current result, not the resulting
JetStream 2 - BrowserBench	Core from website test using Chrome https://browserbench.org/JetStream2.1/
Passmark	CPU Mark Score with Passmark 10 and the newer Passmark Performance Test 11 as well
Crossmark	Benchmark from BapCo using Revision 1.2 documenting the overall score
PCMark 10	Normal test ran, not the express or extended
Procyon	Office productivity benchmark is done with the current Microsoft Office 365, Photo Editing Benchmark is done with Adobe Photoshop and Lightroom Classic.  All products are running on the latest version at the time of testing
Dolphin 5.0 Benchmark	This benchmark runs automatically and at the end gives you a time result in seconds https://bit.ly/2dpa9n3
3DMark	Fire Strike - Physics Score – Performance benchmark, Time Spy - CPU Score, and CPU Profile test
Tom Clancy's Ghost Recon: Wildlands	Built-in Benchmark, 1080p, High detail setting
Team Fortress 2	1080p – high, very high, high, reflect world, high, enabled, 8x msaa, x16, multicore on. TF2 version is locked to previous version that supports older replay files currently until this issue is fixed   https://bit.ly/2vOebin
Dirt 5	Built-in benchmark, running at 1080p on all three resolution settings, V-Sync is turned off, dynamic resolution is turned off, and on the medium detail setting
Far Cry 6	1080p with High Detail setting using the built-in benchmark
Shadow of the Tomb Raider	Built-in benchmark, 1080p, high detail setting, RTX/DLSS Off
Metro Exodus	Default benchmark with High profile which has 1080p and high detail
Watch Dogs Legion	Built in benchmark, 1080p, dx12, low detail
Borderlands 3	Built-in benchmark. fullscreen, 1080p, medium quality, dx11
Power Usage	Idle and load testing using a Kill-A-Watt and wPrime to put the CPU under load and again with the AIDA64 FPU CPU workload
Temps	Aida64 FPU stability test to load and record temperature readings
AIDA64	Memory, Cache, and FLOPS testing done with built-in GPGPU and Memory benchmarks

 

CPU Performance

To start testing off I went with a rendering-focused benchmark with the always popular Blender. I tested with the latest Blender Benchmark available 4.2 and I have all of the results stacked together so we can get the overall picture. I don’t have some of our legacy CPUs tested here, the plan was to also include Blender 4.0 results but both of the new CPUs had problems running that test leaving just these results. This does give us a look at the 285K compared to the 14900K and the 245K next to the 14600K as well. There is a good margin between the 285K and the 14900K but the 245K didn’t pull away as much when compared with the 14600K.

Continuing with the video encoding theme I also have Handbrake which is an open-source transcoder. For this test I am taking a 4k video down to 1080p 30 FPS, the results are the average FPS of that task. The Ultra 9 285K is way out in front here, with a HUGE lead over everything else tested. The Ultra 5 245K also did extremely well, running up near both the 13900K and 14900K in this test.

For the always popular Cinebench, I am testing with the older Cinebench R20 and R23 as well as the new 2024 edition as well. I always like Cinebench because we have multi-core and single-core performance which gives us a good look at the performance of the whole CPU and IPC performance. In R20 on the multi-core test, the Ultra 9 285K is up at the top with a good margin and the Ultra 5 245K was sitting at 9843 vs the 8902 of the 14600K. For the R20 single-core results the 285K was once again up at the top and the 245K moved up into the group of CPUs up at the top. For Cinebench R23 the 285K was once again at the top with plenty of room ahead. I should note however we haven’t gotten the latest Ryzen 9900X and 9950X processors which would also be competing with the 285K. The 285K leads in single-core tests here as well then the 245K is sitting ahead of the 14600K with a healthy lead, almost passing the 7900. Last up was the latest Cinebench 2024 test and its results were similar to R23. What is interesting to see in that test though is the two Ryzen 9000 Series CPUs that we have tested on the single-core results edging out in front of the 245K but the 285K is still way ahead there as well.

Next up we have ray tracing-focused rendering benchmarks. Here I tested the CPUs in POV-Ray and V-Ray including the newer V-Ray Benchmark 5. POV-Ray was also tested with a single core and across all of the cores. The Ultra 9 285K is way out in front for the multi-core test with a score of 14424 and is also in front for the single core with 911.91. The 245K finally pulled away from the 14600K enough to have a few other CPUs sitting between the two and its single-core performance was also impressive in that it caught up but didn’t pass the 13900K and 14900K with their higher clock speeds. In the old V-Ray as well as V-Ray 5 the 245K improved significantly on the performance seen with the 14600K. Interestingly the Ryzen 7950X is still the top result in the V-Ray Benchmark 5 graph which really would be interesting once we get the newest Ryzen CPUs into the mix as well. The 285K did improve significantly compared to the 14900K here, going from 24072 up to 29201.

I also tested using the CPUz’s built-in benchmark which does run on a single thread and with all threads. In the multi-thread test, the 285K improved on the 14900K by over 2000 points. The 245K saw a big jump over the 14600K as well putting it up into the range of the older flagship 12900K. For single-core performance, however, things were more interesting. Both CPUs dropped in performance in this test compared to their previous generation counterparts. The 285K is sitting behind the 14900K with a 906 to the 14900K’s 940 for example.

Next, we have wPrime which is a classic overclocking benchmark that calculates pi out to 1024 million digits and is timed. This is a multi-thread heavy test which has the older high-core count CPUs still all over the top of the charts and you can see that with the top CPUs sitting at 8 or more cores. The Ultra 9 285K scored a 46.071 which was significantly better than the 14900K but it still put it behind the top three Ryzen CPUs from last generation. The Ultra 5 245K scored 99.755 which was much better than the 173.777 of the 14600K but again still below what you would expect. This test has not gone well for any of the Intel CPUs that have e-cores but the improvements we see here are at least going in the right direction.

7 Zip is another open-source program, this time for compressing and decompressing all of your files. Here I have run the benchmark and we have three results. The combined MIPS is a combination of compressing and decompressing performance. Then I have it broken down between the two. The combined results have the Ultra 9 285K again behind the 14900K as well as the 13900K but way out in front of last generation's flagship Ryzen CPUs. The Ultra 5 245K was tied with the 8700G and behind the 14600K by a big margin as well. The decompressing and compressing graph shows us that a lot of the problem was in the decompressing performance, the 285K did outperform the 14900K and 13900K in the compressing performance there but was still behind the 7950X.

Jetstream 2 is a compilation benchmark that takes a long list of HTML5 and Java in-browser tests runs them all three times and puts together an overall score. I love this benchmark because let's be honest, most people are using their browser more than any other game or program. Jetstream gives some interesting results sometimes though but it tends to prefer high IPC or single-core performance and there are some situations where it prefers lower core-count CPUs. The two 9000 Series Ryzen CPUs are still at the top of the chart here but the Ultra 9 285K did at least improve on the performance seen by the 14900K here. The Ultra 5 245K on the other hand did much better with it having a big gap between it and the 14600K and almost matching the 14900K’s performance here.

For Passmark Performance Test I used the latest version, Performance Test 11 and also tested with the older Performance Test 10 as well. I only look at the overall CPU score which takes a few different synthetic benchmarks and combines the results to put together an overall score. This is a test that does favor multi-threaded performance over IPC. In Performance Test 10 the Ultra 9 285K is up at the top and improved on the 14900K’s score by 13.5%. The 245K jumped ahead of the 14600K but there is still a gap between it and the next fastest CPU here. For Passmark Performance Test 11 the Ultra 9 285K is at the top ahead of the 7950X3D here. I was also surprised to see it get the same score across both tests. This time it improved on the 14900K by 13.9% and the 245K bested the 14600K by 10.1%.

Crossmark is from BAPCo which also makes SYSmark and this is a cross-platform test where you can compare performance between phones and both Windows and Mac computers. This is the start of our overall PC benchmarks and Crossmark uses a mix of real-world tests to output an overall score. The 285K is up at the top here once again with a 9.1% improvement over the 14900K. The 245K did less, improving on the 14600K by 3.3%.

PCMark 10 is a great test because it tests things like video calls, browser performance, Excel, and Word performance to give an idea of real-world performance. It tends to like higher clock speeds but does take raw core count into account as well which you can see. The Ultra 9 285K struggled here, coming in right with the last generation 14900K, and with the mid-range Ryzen 7 9700X sitting up at the top it doesn’t give much hope that when we get our hands on the higher-end 9000 Series CPUs that it will even be close. I also test using the PCMark 10 Applications benchmark which uses Microsoft Office for real-world testing and here the 285K did much better, both of the new Core Ultra CPUs are up at the top here with the 245K scoring just ahead of the 14900K.

Moving the focus over towards gaming my next test is using Dolphin 5.0 Benchmark. Dolphin 5.0 is a Wii emulator and like most emulators, it doesn’t care about high core counts at all. It only runs two in total. Clock speeds are king here most of the time which is why all of the 5 GHz+ CPUs are at the top of the chart here. The Ultra 9 285K didn’t outperform the 13900K or 14900K here and has both the 9700X and 9600X from AMD ahead of it as well. The 245K was similar as well finishing the test in 188 seconds to the 183 of the 14600K.

Before diving into game testing I wanted to check out synthetic performance using 3DMark. I tested with the older DX11 Fire Strike test as well as the newer DX12 Time Spy. I also added the new 3DMark CPU Profile benchmark into the mix as well which does a good job of showing the full range of thread counts that you might see being used in games as well as a max threads option that does everything above 16 threads. In the Fire Strike test, the Ultra 9 285K was out in front here. It improved on the 14900K’s score here by 8.7%. The 245K on the other hand just barely improved on what we saw last time with the 14600K with just a 1.4% improvement. For Time Spy however the 285K was down below the 14900K and the 13900K was tied with the 12900K, this was still out in front of what the 7000 Series Ryzen CPUs did. The Ultra 5 245K was sitting ahead of the 7900 and below the 7950X3D but significantly lower than the 14600K. Then in the last test, the CPU profile of both CPUs is out in front of the last generation CPUs in all of the results.

I did of course test the two new CPUs in games as well. I will dive into testing the integrated GPU in the next section. Here though they are paired up with a powerful GPU (the RTX 4090) and I compare CPU to CPU performance. One thing is for sure, there is room for improvement here depending on what games I tested with. Some of the games like Dirt 5 saw a big performance jump coming from the 14900K to the 285K or the 14600K to the 245K but in others, performance dropped. To get a good look at it I put all of the percentages together in the table below. I suspect that there will be more improvements coming to Windows with the APO (application performance optimization) or that our results have an issue. If that is the case I will update this and dive into retesting. Intel did show that their main goals with this generation were similar performance to Raptor Lake-R with power consumption being reduced.

	Ultra 5 245K	Ultra 9 285K
Dirt 5	17.8%	19%
Far Cry 6	0.6%	-2.2%
Shadow of the Tomb Raider	8.5%	2.2%
Metro Exodus	-15.5%	8.6%
Watch Dogs Legion	26.2%	13%
Borderlands 3	-2.3%	-6.6%

 

Last up I also run a few tests in AIDA64, these take a look at cache performance for L1, L2, and L3 cache as well as memory speeds, Precision FLOPS, and Integer IOPS. The biggest thing I wanted to see here was how latency, especially memory latency has changed with the new design. The 14900K came in at 85 ns vs 90 for the 285K which wasn’t as bad as I expected. Single and Double precision FLOPS improved significantly, as did AES-256 performance as well.

AIDA64	L1 Cache Read	L1 Cache Write	L1 Cache Latency	L2 Cache Read	L2 Cache Write	L2 Cache Latency	L3 Cache Read	L3 Cache Write	L3 Cache Latency
	GB/s	GB/s	ns	GB/s	GB/s	ns	GB/s	GB/s	ns
AMD Ryzen R7 2700X	999.43	502.5	1	978.03	495.12	2.8	414.95	417.71	8.8
AMD Ryzen R5 2600X	768.13	388.05	1	729.37	366.01	2.9	429.09	323.83	9.1
Ryzen 5 2400G	485.01	238.29	1	459.8	232.2	3.1	222.96	181.6	9.6
Ryzen 3 2200G	460.55	227.41	1.1	352.84	177.56	3.3	137.3	171.2	10.2
Intel i7-8700K	1595.6	803.65	0.9	640.37	397.14	2.6	337	213.67	11
Intel i5-8400	1417.2	710.79	1	543.81	352.9	3.1	270.89	186.82	13
Intel Core i9-7980XE	3758.3	1896.7	1.1	1671.7	1069.9	5.6	229.61	118.4	21
Intel Core i9-7960X	3532.3	1785.2	1	1571.6	1003.5	5.1	220.22	121.04	19.8
Intel Core i7-7820X	1969.9	995.02	1	884.63	574.74	5.6	114.4	103.57	19.2
Intel Core i5-7640X	1044.8	31888	1	397.26	247	2.9	228.31	163.86	11.1
Intel Core i9-7900X	2418.5	709.16	1	1037.3	521.94	5.2	124.16	106.78	21.2
Intel Core i7-7740X	1102.7	561.25	0.9	424.55	304.3	2.7	287.01	186.67	21.2
Intel i7-6900K	1963.9	999.59	1	710.28	310.34	4.1	243.96	195.41	14.6
Intel i7-7700K	1114.6	560.82	0.9	415.45	285.97	2.7	234.5	188.7	10.4
Intel i7-5960X	1748.5	874.6	1.2	623.08	272.77	3.5	260.72	184.92	15.3
AMD Ryzen R7 1800X	917.46	460.16	1.1	831.95	446.37	4.7	388.18	386.01	12.4
AMD Ryzen R7 1700X	867.89	435.33	1.2	809.29	419.38	5	369.75	350.31	12.9
AMD Ryzen R7 1700	793.9	398.12	1.3	742.92	389.73	5.4	334.36	332.12	14.2
AMD Ryzen R5 1600X	689.47	345.95	1.1	630.99	334.93	4.7	386.48	326.86	12.3
AMD Ryzen R5 1500X	446.84	224.01	1.1	306.82	211.58	11.8	306.82	211.58	11.8
AMD Ryzen R5 2600	685.38	344.05	1	669.28	333.5	3.1	381.16	320.56	9.8
AMD Ryzen R7 2700X	849.91	426.09	1	837.76	418.75	3	357.51	356.71	9.4
Intel i7-8086K	1596.4	803.56	0.9	623.36	423.53	2.6	380.04	243.73	10.9
Intel i9-9900K	2375.6	1195.4	0.8	907.65	626.34	2.4	302.01	221.1	11
AMD Ryzen 9 3900X	3098.6	1551.4	0.9	1545.3	1416.8	2.8	1014.4	836.39	11.5
AMD Ryzen 7 3700X	2133.2	1072.9	0.9	1069.1	993.49	2.8	611.53	573.21	10.2
AMD Ryzen 5 3600X	1547.4	779.24	1	777.62	745.3	2.8	584.93	538.45	10.4
AMD Ryzen 3 3400G	508.97	253.87	1	490.26	247.04	3	233.57	193.62	9.3
AMD Ryzen 3 3300X	1080.5	541.5	0.9	541.84	530.5	2.7	322.13	321.33	11.1
AMD Ryzen 3 3100	965.85	484.67	1	485.55	460.71	3.1	374.11	343.53	11.8
Intel i5-10600K	1676.6	840.87	0.8	662.98	452.67	2.5	361.35	238.47	10.5
Intel i9-10900K	3162.1	1586.7	0.8	1201.8	759.97	2.4	407.66	245.94	11.4
AMD Ryzen 5 5600X	1721.4	869.69	0.9	851.89	831.97	2.6	276.24	298.84	11
Intel i9-11900K	4542.8	2299.3	1	1413.2	691.62	2.5	330.32	213.28	11.4
Intel i5-11600K	3412.3	1724.9	1	960.91	526.76	2.7	319.1	198.57	11.4
Intel i9-12900K	2310	1438.7	1	1233.5	562.72	2.9	678.84	418.11	19.8
Intel i5-12600K	1461.1	971.78	1.1	846.4	360.51	3.2	598.86	292.09	21.4
AMD Ryzen 5 4500	1561.9	784.1	1	785.34	751.61	2.9	350.05	571.19	10.5
AMD Ryzen 5 5600	1646.6	831.24	0.9	832.43	767.22	2.7	445.8	437.73	11.9
AMD Ryzen 7 5700X	2267.8	1157.8	0.9	1147.6	1026.1	2.6	501.53	526.11	11.5
AMD Ryzen 7 7700X	2722.9	1381	0.7	1361.1	1305.1	2.7	887.95	901.2	10.1
AMD Ryzen 5 7600X	1996.3	1019.3	0.7	1020.9	999.93	2.6	847.43	847.61	10.5
AMD Ryzen 9 7950X	5252.2	2685	0.7	2615.6	2588.4	2.6	1293.7	1357.3	9.9
AMD Ryzen 9 7900X	4042.4	2051.4	0.7	2008.5	1956.3	2.5	1278.6	1394.1	9.8
Intel i9-13900K	2005.7	872.39	0.9	2280.2	851.63	3.7	1729.2	619.38	18.3
Intel i5-13600K	3218.4	1472.5	1.1	1156	479.87	4.5	907.63	406.76	14.9
AMD Ryzen 9 7900	3839.2	2023.6	0.7	1925.5	1865.1	2.7	1146	1046.4	10.3
AMD Ryzen 7 7700	2630.8	1333.7	0.8	1336.2	1298.2	2.9	826.97	879.08	10.4
AMD Ryzen 5 7600	1898	963.23	0.8	964.63	946.49	2.9	698.77	690.84	10.8
AMD Ryzen 9 7950X3D	4734.9	2518.4	0.8	2558.7	2422.3	2.7	1314.4	1376.8	12.5
AMD Ryzen 7 7800X3D	2342.7	1239.1	0.8	1252.6	1171.8	3.1	674.02	703.18	12.8
AMD Ryzen 7 8700G	2350.8	1228	0.8	1230.7	1219.8	2.8	867.87	893.95	11.4
AMD Ryzen 5 8600G	1777.2	930.44	0.8	941.76	941.91	2.9	721.29	702.59	10.7
Intel i9-14900K	5923.3	1928.7	0.9	2160.2	861.9	3.6	1158.8	540.95	20.7
Intel i5-14600K	3525.3	1248	1	1310.9	547.83	3.9	590.43	359.64	14.8
AMD Ryzen 5 9600X	3716.7	1942.9	0.7	1429.6	1469.6	2.6	805.86	807.74	11.9
AMD Ryzen 7 9700X	4850.5	2576.9	0.7	1790.2	1804.4	2.5	1078.9	935.36	10.6
Intel Core Ultra 9 285K	5719.9	4012.7	0.7	707.18	711.72	4	2387.3	1261	16.2
Intel Core Ultra 5 245K	3513	2465.1	0.8	574.21	44.44	4.3	1428.4	664.64	15.3

 

 

AIDA64	Memory Read	Memory Write	Memory Latency	Single-Precision FLOPS	Double Precision FLOPS	AES-256	24-bit Integer IOPS	32-bit Integer IOPS	64-bit Integer IOPS
	MB/s	MB/s	ns	GFLOPS	GFLOPS	MB/s	GIOPS	GIOPS	GIOPS
AMD Ryzen R7 2700X	50214	48578	65	507.9	253.9	70514	127.7	127.7	64.22
AMD Ryzen R5 2600X	50273	48346	65.5	383.2	191.4	51932	95.82	95.82	48.15
Ryzen 5 2400G	45967	47850	66.6	235.9	117.8	32388	58.94	58.98	29.67
Ryzen 3 2200G	35567	36240	118.4	225.8	112.6	27603	56.32	56.32	28.3
Intel i7-8700K	38421	40175	57.9	825	412.5	29406	370	370	51.55
Intel i5-8400	38096	40549	59.8	728.8	364.5	25192	323.5	323.5	45.55
Intel Core i9-7980XE	77004	68888	73.4	1831	915.5	69377	821.2	821.3	121.6
Intel Core i9-7960X	79416	71554	74.9	1681	840.6	65409	754	753.9	114.7
Intel Core i7-7820X	76049	73759	71.6	944.9	472.5	36420	423.7	423.7	63.85
Intel Core i5-7640X	30786	31888	68.5	536.5	268.2	18547	238.1	138.1	33.53
Intel Core i9-7900X	76856	72856	72.8	1148	574.5	45506	515.1	515	79.76
Intel Core i7-7740X	38455	40596	53.2	575.4	287.8	20503	258.1	258	35.97
Intel i7-6900K	66786	68130	61.9	1023	511.7	36340	255.9	255.9	63.96
Intel i7-7700K	38498	40448	52.7	575.7	287.9	20519	258	258	35.98
Intel i7-5960X	66108	49545	62.8	895.5	447.7	31783	223.9	223.9	55.97
AMD Ryzen R7 1800X	44031	43425	81.8	472.4	236.2	64016	118.1	118.1	59.03
AMD Ryzen R7 1700X	44493	43749	78.3	446.8	223.4	61730	111.7	111.7	55.84
AMD Ryzen R7 1700	45343	43777	82.6	408.5	204.3	56811	102.1	102.1	51.05
AMD Ryzen R5 1600X	44452	43918	83.7	354.9	177.5	49340	88.74	88.73	44.34
AMD Ryzen R5 1500X	44289	43746	83.1	229.8	114.8	31983	57.43	57.45	28.72
AMD Ryzen R5 2600	50385	48499	66.5	351	175.4	48939	88.03	88.02	44.01
AMD Ryzen R7 2700	50088	48343	65.9	434.3	217.2	60336	108.6	108.6	54.68
Intel i7-8086K	39069	40675	56.2	824.9	412.5	29397	369.9	369.9	51.55
Intel i9-9900K	39682	41015	51.1	1226	613	45542	550	549.8	79.79
AMD Ryzen 9 3900X	53981	52271	70.7	1546	772.1	106943	385.7	385.7	96.57
AMD Ryzen 7 3700X	51498	28739	67.3	1078	537.8	74474	271	271.2	67.77
AMD Ryzen 5 3600X	50769	28736	68.9	799.5	400	55456	198.6	198.6	49.64
AMD Ryzen 3 3400G	49936	53502	67.1	257.2	128.5	35736	64.16	64.16	32.25
AMD Ryzen 3 3300X	51435	28743	66.4	551.6	276.1	38176	138.7	138.7	34.65
AMD Ryzen 3 3100	50246	28734	71.1	497.6	248.7	34946	124.4	124.4	31.09
Intel i5-10600K	43340	43430	48.2	863.8	431.9	30782	387.2	386.9	53.99
Intel i9-10900K	44972	43808	50.5	1632	815.7	58163	731.3	731.2	102
AMD Ryzen 5 5600X	50972	28772	59.2	891.9	445.8	61623	709.6	709.7	55.72
Intel i9-11900K	44919	43343	51.8	1068	534.2	164966	547	547	182.7
Intel i5-11600K	43953	43468	55.6	838.7	419.4	123767	429.3	429.3	143.4
Intel i9-12900K	71418	72395	82	695.7	247.9	140499	400.6	400.6	166.2
Intel i5-12600K	65416	71336	87.8	432.5	216.2	86470	252.7	252.7	96.22
AMD Ryzen 5 4500	52683	53191	82.8	806	402.6	56578	201.5	201.3	50.35
AMD Ryzen 5 5600	48444	28753	70.9	853.2	426.2	107839	675.8	679.6	53.38
AMD Ryzen 7 5700X	47858	28760	68.2	1158	577.9	137459	919.1	912.9	74.38
AMD Ryzen 7 7700X	58755	80120	70	1392	694.6	192730	1205	1205	602.4
AMD Ryzen 5 7600X	54720	80771	69.5	1046	523.1	147661	912.5	912.5	456.4
AMD Ryzen 9 7950X	67342	76712	69	2735	1364	379510	2370	2370	1184
AMD Ryzen 9 7900X	68064	78123	67.7	2062	1029	288134	1792	1789	895.4
Intel i9-13900K	76006	68902	86.1	2238	1116	157028	831.4	831.4	270.6
Intel i5-13600K	64721	62311	97.2	1294	647	86637	509.5	509.4	136.4
AMD Ryzen 9 7900	73968	73494	69.4	2007	998.7	267499	1710	1709	855.9
AMD Ryzen 7 7700	58694	79929	68.9	1369	684.7	185084	1192	1192	594.3
AMD Ryzen 5 7600	59454	81467	69.6	988.6	494.3	140625	861.9	861.2	431.2
AMD Ryzen 9 7950X3D	77677	80314	68.1	2619	1311	178269	2239	2239	1120
AMD Ryzen 7 7800X3D	65343	81313	71.9	1241	620.3	89694	1076	1076	538.3
AMD Ryzen 7 8700G	75149	99082	74.5	1277	638.1	88895	1108	1108	554.1
AMD Ryzen 5 8600G	75418	99134	71.6	969.3	484.6	69303	842.7	843.2	421.4
Intel i9-14900K	85919	76896	85.5	2462	1231	163454	936.6	937.3	291.6
Intel i5-14600K	79924	77188	83.9	1407	703.3	98868	560.1	560	154
AMD Ryzen 5 9600X	62215	84220	68.7	2039	1016	64063	1753	1753	876.8
AMD Ryzen 7 9700X	62204	84918	68.7	2683	1337	83891	2297	2292	1147
Intel Core Ultra 9 285K	94135	85518	90.3	3553	1756	248079	1355	1413	440.9
Intel Core Ultra 5 245K	89004	85829	89.6	2103	1045	149961	784.9	786.1	267.6

 

 

Onboard Video Performance

For most people, CPU performance is the most important aspect when picking out a processor. But some people do still run without a dedicated GPU so I still take a look at the performance of the integrated GPUs as well. In the case of Intel’s new CPUs have changed things up when compared to the 14^th-generation CPUs. They have the Xe-LPG with four Xe cores. This is the same iGPU as Meteor Lake CPUs aka the Core Ultra 200 Series which was found in mobile and IoT and embedded systems. They didn’t go with the Battlemage Xe2 engine iGPUs that are found on the Lunar Lake mobile CPUs. For clock speed, the Core Ultra 9 285K has a 2 GHz clock speed and the Core Ultra 5 245K drops that down to 1.9 GHz but otherwise, they are the same. Intel’s onboard graphics for their desktop processors have been the same for a long time so I’m excited to see some movement. Especially with AMD making significant progress with their 8000G series CPUs. To start things off I tested with 3Dmark. I ran tests using the older Fire Strik benchmark, the DX12-based Time Spy, the newer Speedway test, and the latest test Steel Nomad which for that last one I tested using the light version. Both the Ultra 9 285K and the Ultra 5 245K had significant jumps in performance over the 14900K and the 14600K. Each doubles the previous score or better. This wasn’t enough to catch up to the 8700K or 8600G though. As for that 1GHz clock speed difference as well as the differences between the CPUs themselves, the 285K is out in front of the 245K by 3-6 percent depending on the test.

In Unigine Superposition I ran the 1080p medium detail and 720p low detail tests and the 285K and 245K didn’t double the performance on the 720p low tests but was closer on the 1080p medium test. Once again they were both sitting behind the 8700G and 8600G by a good amount but well in front of everything else including Intel’s 14^th Gen CPUs.

Then from there on, I jumped into game tests. Some of our tests are older games but I did also add in a few newer games as well to get an idea of newer games when tested at 1080p and low or medium settings. My goal with these tests was to see if base-level gaming at low or medium settings was possible at all. In four of the 6 games tested both of the Intel Core Ultra CPUs doubled up the performance seen with the 14^th gen CPUs. For the Hitman Absolution and Tomb Raider results there was still a large improvement but not reaching twice the frame rate, 75% and 56% respectively. The Ryzen 8000G CPUs were still out in front of course but for me, the big news here was the improved performance pushing two of the games up and over 60 FPS for smooth gameplay and the other four went from being unplayable to playable but not smooth in the 40 FPS range. Obviously, a dedicated GPU is still the way to go, but if you need to get by while you save up or wait for your GPU to come in you might just be able to get by on the lowest settings in some games.

 

 

Power Usage and Temperatures

For some people, performance is the only thing important, but for others, power usage and temperatures also play a role so we do take a look at both of those as well. This is especially important in SFF or even just smaller mid-sized builds and it affects the components you need to get for your system as well as your PSU and cooler. To take a look at power usage I ran three different tests. I noted the idle power draw of our entire system then I took a look at the load wattage of the system using two different workloads. One was wPrime and the second was AIDA64 using their FPU workload which is extremely demanding. At idle the testbench with the Ultra 9 285K pulled 77.1 watts and with the Ultra 5 245K, it was down to 69.2 watts. This was better than the two 9000 Series Ryzen CPUs I tested and the last generations 14900K and 14600K. Idle and near idle is where a lot of usage is and it's good to see those numbers dropping back down. Under load, the Ultra 9 285K pulled 252 watts when using wPrime and 328 watts with the FPU workload putting the CPU to work. With the AIDA64 workload that was 17.8% less than with the 14900K in that same test. The 245K pulled 152 watts in wPrime and 223 with the Aida64 FPU workload. It tied the 14600K in that specific test which was interesting.

I’ve spoken in the past about how temperature testing isn’t an end-all-be-all-be-all result. CPU to CPU with the same CPU can be different and that gets even more complicated once you add in different motherboards and BIOS revisions as well. Not to mention different coolers. In this case, we know that Intel’s new design has moved things around on the chip. Older coolers work, but it looks like there are companies with coolers better optimized for LGA 1851 are coming as well so cooler performance also contributes to these numbers as well. That said the Ultra 9 285K did improve on the temperatures that I was seeing when running this same test on the 14900K. The 14900K also ran into thermal throttling where the 285K didn’t but that isn’t to say this CPU runs cool, at least not with this workload. Intel’s focus is improving temperatures while gaming and this is a significantly higher load than that. For comparison the 285K ran at 91c with the FPU workload but at 72c with the AIDA64 CPU workload which is closer to what you would see when gaming. The Ultra 5 245K also improved as well dropping from 77c down to 64c.

 

Overall and Final Verdict

With all of our testing done (other than iGPU testing which is still coming) we can step back and get a better look at the overall picture with both of Intel’s new Core Ultra 200 Series CPUs. Sadly I don’t have the Ryzen 9900X and 9950X to see how the Ultra 8 285K compares there. But looking at how the 285K compares with last generation’s Core i9-14900K tells us a lot. In a lot of our tests, it shows a significant improvement over the last generation, especially in some of our real-world tests. Intel said their goal was to improve thermals and power efficiency while getting similar gaming performance as the last generation. They did accomplish most of that. Both the 285K and the 245K pulled less power and with that ran cooler than their 14^th Gen Core counterparts. But the other half of the story is just how hit-and-miss performance is, especially when it comes to gaming. In some games, I saw 10%-25% performance improvements, but in other games, both CPUs performed worse than their last-gen counterpart. It is very clear that we are seeing performance issues and I wouldn’t be surprised if we see significant improvements over the life of these CPUs as we get driver/windows/BIOS updates. It rarely feels like Intel has pushed out CPUs before they were ready, but that seems to be the case this time around. We could speculate that might be in response to the issues with their 13^th and 14^th gen CPUs even though they have released BIOS updates to help prevent those issues.

How did the Ultra 9 285K come out on pricing? Well, it has the same launch MSRP as the i9-14900K did with a suggested price of $589. The Ultra 5 245K on the other hand comes in $10 cheaper than the 14600K with a suggested price of $309. I’m happy to see the price come down slightly there. Especially with the Ryzen 9700X being about $20 more, when it comes to non-gaming performance the Ultra 5 245K outperformed the 9700X in a majority of our tests making it a good option for those situations. Of course for gaming, the inconsistency is an issue and I hope we see that improve as the platform matures.

Live Pricing: HERE

Live Pricing: HERE

• LGA1851
• Intel
• Core Ultra
• 14 Cores
• 24 Cores