Today, we tin yet return nan covers disconnected Intel's latest CPU generation, nan "Core Ultra" series, codenamed Arrow Lake. The Arrow Lake architecture is caller for desktop systems, utilizing a tile-based processor design, pinch nan extremity of moving distant from a ample monolithic spot based connected nan latest node and alternatively splitting nan CPU into blocks, aliases "tiles," arsenic they're commonly referred to.
This attack allows Intel to create nan CPU cores and iGPU utilizing cutting-edge nodes, while different aspects of nan processor, specified arsenic I/O interfaces, tin beryllium built utilizing an older, much cost-effective node. This attack is akin to what AMD has been doing pinch Ryzen.
Since Arrow Lake represents specified a extremist alteration from nan erstwhile Raptor Lake generation, we could walk a batch of clip discussing nan architectural and level changes, immoderate of which we've been covering complete clip successful anticipation for this release. However, since astir of you are present for nan benchmarks and existent capacity results, we'll quickly spell complete nan flagship CPU we're testing today, touch connected nan caller platform, and past dive into nan graphs.
Meet nan caller Core Ultra 9 285K
The caller flagship is nan Core Ultra 9 285K. Yes, it's not nan champion name, and yes, nan 'Ultra' is unnecessary, but Intel is conscionable pursuing nan inclination – modern merchandise names thin to beryllium unnecessarily agelong and somewhat uninspired, truthful that container is checked.
The 285K features 8 P-Cores pinch 8 threads, arsenic Hyper-Threading is nary longer used. These P-Cores person a guidelines wave of 3.7 GHz, boosting up to 5.6 GHz, pinch a thermal velocity boost of 5.7 GHz – resulting successful a 5% wave simplification compared to nan 14900K.
There are besides 16 E-Cores, each pinch 1 thread, arsenic location is nary SMT support here. These E-Cores run astatine a guidelines wave of 3.2 GHz, pinch a boost up to 4.6 GHz, which is simply a 5% summation complete nan E-Cores successful nan 14900K.
Cores (threads) | Freq. | Turbo | L2 cache | Cores (threads) | Freq. | Turbo | L2 cache | |||
Core Ultra 9 285K | $600 | 8 (8) | 3.7 GHz | 5.6 GHz | 24 MB | 16 (16) | 3.2 GHz | 4.6 GHz | 16 MB | 36 MB |
Core Ultra 7 265K | $395 | 8 (8) | 3.9 GHz | 5.5 GHz | 12 (12) | 3.3 GHz | 12 MB | 30 MB | ||
Core Ultra 5 245K | $310 | 6 (6) | 4.2 GHz | 5.2 GHz | 18 MB | 8 (8) | 3.6 GHz | 8 MB | 24 MB | |
Core i9-14900K | $600 | 8 (16) | 3.2 GHz | 6.0 GHz | 16 MB | 16 (16) | 2.4 GHz | 4.4 GHz | 16 MB | 36 MB |
Core i7-14700K | $420 | 8 (16) | 3.4 GHz | 5.6 GHz | 12 (12) | 2.5 GHz | 4.3 GHz | 12 MB | 33 MB | |
Core i5-14600K | $330 | 6 (12) | 3.5 GHz | 5.3 GHz | 12 MB | 8 (8) | 2.6 GHz | 4.0 GHz | 8 MB | 24 MB |
In total, location are 36 MB of L3 cache and 40 MB of L2 cache. Each P-Core gets 3 MB of L2 cache, while nan E-Cores are grouped into clusters of four, pinch each cluster sharing 4 MB of L2 cache. Finally, nan guidelines TDP is 125W, pinch a max turbo TDP of 250W. Intel is pricing this astatine $590 per 1,000 units, truthful we expect an first unit value of astatine slightest $600.
It's besides worthy noting that each K-SKU models, which are nan only ones announced truthful far, support dual-channel DDR5-5600 UDIMM representation aliases DDR5-6400 CUDIMM memory. CUDIMM representation includes a mini timepiece driver circuit straight connected nan module, enabling much precise timings required astatine higher representation speeds.
All CPUs supply 20 PCIe 5.0 lanes and 4 PCIe 4.0 lanes, on pinch an 8-lane Direct Media Interface 4.0 autobus to nan chipset. As usual, nan K-suffix processors person an unlocked multiplier, making them perfect for overclocking.
A New Socket
With Arrow Lake, Intel has moved distant from nan LGA1700 socket, switching to LGA1851. Despite nan accrued pin count, nan caller socket maintains nan aforesaid dimensions and cooler mounting spread spacing arsenic LGA1700, ensuring continued compatibility pinch existing CPU coolers.
For testing, we utilized a number of caller motherboards, chiefly nan Asus ROG Maximus Z890 Hero and MSI MEG Z890 Unify-X. A important number of BIOS updates were needed to execute nan capacity figures we're astir to present, truthful it's adjacent to opportunity nan Arrow Lake reappraisal process has been acold from smooth.
Testing Notes
First, to beryllium transparent, we had very constricted clip to hole this review, arsenic Intel only sent america nan reappraisal kit 3 days earlier nan embargo lifted. Three days is not capable clip to tally each nan tests we'd for illustration to, truthful immoderate tests will beryllium covered successful follow-up content.
Since these CPUs will beryllium disposable for acquisition connected October 24th, we wanted to person a reappraisal fresh to thief guideline your buying decisions. However, delays successful receiving reappraisal samples turned retired to beryllium nan slightest of our concerns, arsenic Arrow Lake is presently a spot of a basking mess, truthful overmuch truthful that we're not judge wherever to start.
Let's commencement pinch Windows 11. Many of you cognize that Windows 11 24H2 was released a fewer weeks ago, importantly improving gaming capacity for modern Ryzen processors successful immoderate cases. It moreover benefits 13th and 14th-gen Intel models. However, erstwhile updating our data, we noticed that 12th-gen Intel CPUs performed noticeably worse connected 24H2 compared to 23H2.
This rumor wasn't constricted to 12th-gen processors. We encountered superior capacity and stableness problems pinch Arrow Lake connected 24H2. After aggregate caller installs grounded to resoluteness nan issue, we had to revert to testing connected 23H2. This is little than ideal, arsenic we cognize immoderate games execute worse connected nan older type of Windows 11.
Even pinch a caller 23H2 install, we still knowledgeable stableness issues pinch nan 285K. Games would occasionally crash, arsenic would definite applications. The problem was worse erstwhile utilizing nan DDR5-8200 representation provided for testing. This persisted moreover aft galore BIOS revisions for 2 different motherboards.
We expect these teething issues will beryllium addressed quickly, though we are unsure why 24H2 caused truthful galore problems. It's peculiarly overseas because Intel reportedly did each their benchmarking connected 24H2 utilizing nan Insider Preview build from August and did not study immoderate issues. Nevertheless, we've spoken pinch aggregate reviewers who besides encountered problems, moreover pinch 23H2, and committee manufacturers confirmed location are issues pinch 24H2.
Test System Specs
Before we get into nan benchmark data, present are nan trial specs and systems used. One point to statement is that we benchmarked nan 285K pinch some DDR5-7200 and DDR5-8200 memory.
CPU | Motherboard | Memory |
AMD Ryzen 7000 Series | Gigabyte X670E Aorus Master [BIOS F33d] | G.Skill Trident Z5 RGB 32GB DDR5-6000 CL30-38-38-96 Windows 11 24H2 |
AMD Ryzen 5000 Series | MSI MPG X570S Carbon MAX WiFi [BIOS 7D52v19] | G.Skill Ripjaws V Series 32GB DDR4-3600 CL14-15-15-35 Windows 11 24H2 |
Intel Core Ultra 200S | Asus ROG Maximus Z890 Hero [BIOS 0805] | G.Skill Trident Z5 CK 32GB DDR5-8200 CL40-52-52-131 G.Skill Trident Z5 RGB 32GB DDR5-7200 CL34-45-45-115 Windows 11 23H2 [24H2 = Slower] |
Intel 12th, 13th & 14th | MSI MPG Z790 Carbon WiFi [BIOS 7D89v1E] | G.Skill Trident Z5 RGB 32GB DDR5-7200 CL34-45-45-115 Windows 11 23H2 [24H2 = Slower] |
Graphics Card | Power Supply | Storage |
Asus ROG Strix RTX 4090 OC Edition | Kolink Regulator Gold ATX 3.0 1200W | TeamGroup T-Force Cardea A440 M.2 PCIe Gen4 NVMe SSD 4TB |
GeForce Game Ready Driver 565.90 WHQL |
The 7200 CL34 representation runs pinch overmuch tighter timings compared to nan DDR5-8200 CL40 memory. If an exertion aliases crippled is much latency-sensitive, nan 7200 representation whitethorn execute better. However, if bandwidth is nan limiting factor, nan 8200 kit mightiness person nan advantage.
Now, let's spot really nan numbers shingle out...
Productivity Benchmarks
Cinebench Clock Behavior
First, let's return a look astatine really nan 285K behaves nether load. For cooling, we are utilizing nan caller MSI MAG Coreliquid I360, which was provided successful our reappraisal kit and is specifically designed and optimized for nan caller Arrow Lake CPUs.
MSI has developed a unsocial bracket that shifts nan acold sheet north, resulting successful a 3-degree simplification successful somesthesia arsenic it amended targets nan hotspot connected these caller CPUs. A bully characteristic of this creation is that nan unsocial mounting bracket besides provides optimal sum for LGA1700, AM4, and moreover AM5 processors, making it a versatile, one-size-fits-all solution.
With nan Coreliquid I360 installed, we loaded up nan 285K pinch Cinebench and observed an mean timepiece wave of 4.6 GHz connected nan E-Cores and 5.3 GHz connected nan P-Cores, each while staying wrong nan banal 250W powerfulness limit. The CPU reached a highest halfway somesthesia of 84°C, which is good beneath nan 105°C TjMAX.
Cinebench 2024, Multi Core
Now for nan benchmark graphs: nan Cinebench multi-core capacity looks very strong. We already cognize that E-Cores activity good successful this type of workload, based connected erstwhile models. The E-Cores successful nan 285K are based connected an updated architecture and tally astatine somewhat higher timepiece speeds. Using either DDR5-7200 aliases 8200 memory, we saw a people of conscionable complete 2,500 points, making nan 285K 14% faster than nan 14900K and 7% faster than nan 9950X.
Cinebench 2024, Single Core
When we look astatine single-core performance, nan Arrow Lake P-Cores proceed to impress, pinch nan 285K being 13% faster than nan 14900K and 6% faster than nan 9950X – different awesome result.
Cinebench 2024 Power Consumption
Next, let's talk powerfulness consumption. This portion of nan reappraisal gave america immoderate trouble, but acknowledgment to Steve from Gamers Nexus, we were capable to drawback an rumor pinch nan testing. The Asus ROG Maximus Z890 Hero has a unsocial design: alternatively of delivering each nan CPU powerfulness done nan dual EPS12V rails, 4 of nan vCore powerfulness stages are connected to nan 24-pin ATX powerfulness cable.
Initially, we were puzzled arsenic to why nan 285K, pinch a 250W limit, was drafting only 196W. After Gamers Nexus alerted america to this, we confirmed that astir 50-60W of powerfulness was being supplied via nan 24-pin ATX cable. This was later confirmed by Asus engineers.
After retesting pinch nan MSI MEG Z890 Unify-X, we confirmed nan correct EPS12V data, showing nan 285K drafting 258W during this test. While this is an betterment complete nan 14900K, it's not peculiarly impressive, considering nan 9950X was only 7% slower but consumed 11% little power. Even much concerning is that nan 7950X3D was conscionable 13% slower than nan 285K but utilized a monolithic 43% little power.
However, compared to nan highly inefficient 14900K, nan 285K represents a measurement successful nan correct direction, consuming 17% little powerfulness while delivering 14% much performance.
7-Zip File Manager Compression
When it comes to compression performance, nan 285K is solid, but not groundbreaking. It performs likewise to nan 14900K, 9950X, and 7950X. While it's not a immense leap guardant for a caller generation, it's surely not a bad result.
7-Zip File Manager Decompression
However, decompression capacity is rather weak, mostly owed to nan removal of SMT. As a result, erstwhile utilizing nan aforesaid wave memory, nan 285K is 9% slower than nan 14900K and a important 25% slower than nan 9950X.
Blender Open Data
The Blender Open Data results are good, though not exceptional, arsenic nan 285K matches nan 9950X. However, fixed nan improved powerfulness efficiency, this is still a affirmative result overall.
Corona 10 Benchmark
In nan Corona 10 Benchmark, nan 285K performs well, though it's 9% slower than nan 9950X, contempt showing a 17% capacity summation complete nan 14900K.
Adobe Photoshop 2025
Despite nan awesome single-core scores seen successful Cinebench, nan 285K's capacity successful nan lightly threaded Photoshop benchmark is disappointing. It's somewhat slower than nan Core i5-12600K, making it 1 of nan slowest CPUs we've tested successful this benchmark, which is some puzzling and disappointing.
Adobe Premiere Pro 2024
Fortunately, capacity successful Adobe Premiere Pro is overmuch better, though still somewhat down nan 14900K. This result, while better, is still underwhelming for a next-generation product.
Gaming Benchmarks
Let's dive into nan gaming benchmarks, starting pinch Star Wars Jedi: Survivor. Ryzen processors received a notable capacity boost successful this title pinch Windows 11 24H2. While nan gains for 14th-gen Intel processors were smaller, they were still present. Unfortunately, parts for illustration nan 285K are presently astir 5-10% slower connected 24H2, truthful nan 285K information presented present was gathered utilizing 23H2.
Star Wars Jedi: Survivor
Ignoring nan X3D CPUs, which are blazing accelerated successful this test, nan Intel 285K delivered reasonably mid-range performance, coming successful 4% slower than nan 14900K. Additionally, nan lower-latency DDR5-7200 representation was somewhat faster than nan 8200 representation successful this case.
The Last of Us Part 1
The 285K performed impressively successful The Last of Us Part 1, delivering capacity comparable to nan 7800X3D and somewhat faster than nan 14900K. However, pinch nan aforesaid 7200 memory, we only observed a 2% improvement, which is underwhelming for a next-gen product. Some mightiness moreover telephone it a disappointment.
Cyberpunk 2077: Phantom Liberty
This is wherever things really spell disconnected nan rails. You mightiness spot reviews wherever nan 285K performs likewise to nan 14900K successful Cyberpunk 2077: Phantom Liberty, but those reviews apt utilized nan built-in benchmark aliases tested sections of nan crippled that aren't very CPU-intensive. We person grounds for this, which we will explicate later.
In much CPU-demanding sections of nan game, nan 285K struggles significantly. We've spent sizeable clip attempting to extract much capacity successful this test, pinch nan champion results coming from lower-latency memory, wherever tightening timings helped nan most. However, contempt testing motherboards from MSI, Gigabyte, Asrock, and Asus, and moreover performing aggregate caller Windows installs, nan 285K remained astatine slightest 20% slower than nan 14900K, making it nan slowest CPU successful this test.
Hogwarts Legacy
Performance successful Hogwarts Legacy is better, but still mediocre overall. The 285K lagged down nan 14900K, making it slower than moreover nan Ryzen 7 7700X. It's astonishing to spot really overmuch Windows 11 23H2 was holding backmost Zen 4 and Zen 5 processors, arsenic we now regularly find nan 7700X matching nan 14900K's performance. Additionally, nan Intel CPU is nary longer moving without powerfulness limits.
Assetto Corsa Competizione
In ACC, nan 285K fell down nan 14900K by 5%, making it 14% slower than nan 9700X and 32% slower than nan 7950X3D, meaning Arrow Lake is not an perfect prime for this driving simulator.
Remnant II
Remnant II produced information person to what Intel had promised, pinch nan 1% lows being importantly amended connected nan 285K compared to nan 14900K. However, contempt matching Intel's erstwhile flagship, nan 285K still trailed down nan Zen 4 X3D parts.
Homeworld 3
Another crippled wherever nan 285K's capacity falls to nan bottommost of nan floor plan is Homeworld 3. In this test, nan 285K delivered Core i5-12600K-like performance, making it almost 20% slower than nan 14900K and 7950X3D, and astatine slightest 23% slower than nan 9700X.
A Plague Tale: Requiem
Performance successful A Plague Tale: Requiem is besides shocking. If you've seen different reviews showing overmuch amended results successful this title, we'll explicate nan discrepancy later. For now, nan 285K is underwhelming, allowing for conscionable complete 120 fps connected average, pinch 1% lows successful nan mid-70s. It's astatine slightest 17% slower than nan 14900K and 27% slower than nan 7950X3D, again delivering Core i5-12600K-like performance, making it moreover slower than nan Ryzen 5 7600X.
Counter-Strike 2
In Counter-Strike 2, nan results align much intimately pinch Intel's promises, pinch nan 285K astir matching nan 14900K. While it's bully to spot this benignant of parity, it's acold from nan norm based connected our different tests, and still an underwhelming consequence overall.
Starfield
Interestingly, nan 285K delivered its champion capacity successful Starfield, narrowly beating nan 14900K and moreover surpassing nan 7950X3D. However, utilizing nan aforesaid memory, it was only 2% faster than nan 14900K, which, while 1 of nan amended results we've seen, is still disappointing for a caller generation.
Warhammer 40,000: Space Marine 2
Space Marine 2 ran good connected nan 285K, though this caller flagship only offered mid-tier performance, coming successful slower than nan 14700K and astir matching nan Ryzen 7 7700X. It's not a disaster, but acold from impressive.
Hitman 3
The 285K performed decently successful Hitman 3, being 1 of nan fastest CPUs successful this title. However, it was still somewhat slower than nan 14900K – by conscionable 2% – truthful it's difficult to get excessively excited astir this result, arsenic we're looking astatine 14700K-like performance.
Watch Dogs: Legion
For testing Watch Dogs: Legion, we utilized nan built-in benchmark, but averaged nan information from six runs, arsenic nan first results are often inflated. To accurately comparison pinch this data, you'll request to return an mean of six back-to-back runs.
Doing truthful reveals highly disappointing results for nan 285K, which only matched nan 14600K and 9600X pinch 162 fps connected average. This made it 11% slower than nan 14900K and 20% slower than nan 7950X3D.
Star Wars Outlaws
Finally, successful Star Wars Outlaws, nan lower-latency DDR5-7200 representation provided nan champion results, pinch 134 fps – nan aforesaid capacity level arsenic nan 12900K. And pinch that, we reason our highly disappointing gaming benchmarks.
Side-by-Side Data
To recap, here's really nan 14900K and 285K compared, some utilizing DDR5-7200 memory. On average, nan 285K was 6% slower crossed nan 14 games we tested, pinch important losses successful Watch Dogs: Legion, A Plague Tale: Requiem, Homeworld 3, and Cyberpunk 2077: Phantom Liberty.
Gaming Power Consumption
Arrow Lake makes important strides successful powerfulness efficiency, and while it's overmuch improved complete nan 14900K, nan results still autumn short erstwhile compared to Ryzen processors.
In our Cyberpunk testing, we observed a important 32% simplification successful powerfulness usage for nan 285K compared to nan 14900K. However, it still consumed 28% much powerfulness than nan 9950X, and capacity for nan 285K was mediocre successful this title. Let's now look astatine a crippled wherever nan 285K performed better.
In The Last of Us Part 1, nan 285K consumed 32% little powerfulness than nan 14900K, which is awesome fixed it was a fewer percent faster. Unfortunately, erstwhile compared to nan 7950X3D, nan 285K utilized 71% much powerfulness for nan aforesaid level of performance, which is disappointing. While Intel has made improvements successful powerfulness efficiency, reducing powerfulness depletion by almost 70W from nan 14900K, it still lags down AMD, particularly astatine nan existent capacity levels.
14 Game Average
Looking astatine nan 14-game average, we spot that nan 285K is 5% slower than nan 14900K, erstwhile calculating utilizing nan geometric mean. This intends that, connected average, it's somewhat slower than nan 7700X and astir connected par pinch nan 9600X, which is rather surprising. It's besides 14% slower than nan 7950X3D.
Here, we spot conscionable really overmuch Ryzen processors person benefited from nan 24H2 update, pinch nan 7800X3D being 14% faster than nan 14900K, which is remarkable. Of course, we are now power-limiting nan Core i9, but moreover so, that separator is impressive.
Cost Per Frame
Typically, our cost-per-frame study for CPUs includes nan costs of nan motherboard and memory, but successful this case, nan prices for those components are rather similar. While this whitethorn trim nan wide margins slightly, nan image is clear: nan 285K offers mediocre worth for gamers.
It provides akin worth to nan 9950X, making it overmuch worse than nan 7950X3D, which is astir apt nan champion all-rounder astatine nan moment. It besides performs worse than nan outgoing 14900K, which we person ne'er recommended.
Interestingly, nan 7800X3D, which has sold for arsenic small arsenic $340 this year, still offers acold amended worth astatine $480. Even pinch caller value increases, nan 7800X3D remains a beardown contender successful position of value.
Windows 24H2 Gaming Performance
As mentioned earlier, Arrow Lake's gaming capacity was moreover worse connected Windows 11 24H2. Intel is alert of this rumor and has suggested enabling nan 'high performance' aliases 'high power' floor plan successful Windows for optimal capacity connected 24H2. They besides speculated that location mightiness beryllium an rumor pinch nan ringing autobus timepiece connected 24H2, suggesting a BIOS alteration by disabling nan 'CCF Auto GV' setting.
We explored these suggestions utilizing a constricted group of gaming benchmarks to spot nan impact. Additionally, we included a caller instal of Windows 10, arsenic galore users still trust connected this OS and whitethorn beryllium funny astir nan 285K's capacity connected it.
Star Wars Jedi: Survivor
Looking astatine nan 23H2 data, nan 285K's capacity improved by 3% erstwhile utilizing Windows 10, which is simply a bully sign. However, it's apt that parts for illustration nan 14900K would spot a akin aliases perchance greater capacity uplift.
Using nan default 24H2 configuration pinch nan balanced powerfulness floor plan severely hindered nan 285K, reducing nan mean framework complaint by 8% compared to 23H2, pinch nan 1% lows dropping by 36%. Clearly, thing is incorrect here. Switching to nan 'high performance' powerfulness scheme improved nan framework complaint to 150 fps connected average, though still beneath what we saw pinch 23H2, but a notable betterment nonetheless.
Cyberpunk 2077: Phantom Liberty
In Cyberpunk 2077: Phantom Liberty, nan 285K performed champion utilizing 23H2, pinch a flimsy framework driblet erstwhile utilizing Windows 10, though nan quality was wrong nan separator of error.
Interestingly, nan 24H2 balanced powerfulness scheme performed amended here, arsenic nan framework complaint regressed by 3% erstwhile switching to nan precocious capacity profile. Adjusting nan 'CCF Auto GV' mounting had nary impact, akin to what we saw successful Star Wars Jedi: Survivor.
Homeworld 3
The Homeworld 3 information underscores conscionable really erratic nan Arrow Lake testing has been, though nan results align pinch expectations. We cognize Homeworld 3 performs poorly connected 23H2, while 24H2 importantly boosts capacity for Intel's 13th and 14th-gen and AMD Ryzen processors. The truth that Windows 10 performed 20% amended than 23H2 makes consciousness successful this context.
What doesn't make consciousness initially is nan deficiency of betterment connected 24H2 for nan 285K – until you manually set nan powerfulness plan. After doing so, nan capacity matched nan Core i7-14700K.
Unfortunately for Intel, we person to take a azygous OS configuration for testing, and location seems to beryllium nary cleanable action wherever everything useful arsenic it should, astatine slightest not astatine this time.
A Plague Tale: Requiem
Performance successful A Plague Tale: Requiem was disappointing connected nan 285K, and switching to 24H2 didn't resoluteness nan issue. Oddly, Windows 10 performed poorly successful this title, though this was a inclination we observed successful a fewer of nan 14 games tested.
Counter-Strike 2
In Counter-Strike 2, nan capacity connected 23H2 matched what we saw pinch Windows 10. However, 24H2 was consistently slower, though switching to nan precocious capacity powerfulness floor plan provided a mini boost.
Intel APO Optimization
Lastly, it's important to statement that each nan gaming benchmarks you conscionable saw were tally without Intel's APO (Application Optimization) characteristic enabled. This characteristic was not enabled by default during our testing, but Intel has indicated that this will beryllium addressed successful early BIOS updates.
We didn't attraction connected this for 2 reasons. First, retired of nan 14 games we tested, only 2 – Counter-Strike 2 and Cyberpunk 2077 – support APO. Second, retired of curiosity, we went backmost and enabled APO successful those games and recovered only a 1-2% alteration successful performance, which falls wrong nan separator of error.
This suggests that either APO has minimal effect connected those titles, aliases it isn't afloat functional yet. We will revisit this characteristic successful early contented to spot if location are immoderate important updates aliases improvements.
What We Learned: Inconsistent Gaming Performance
So, location we person it – a chaotic operation of results that often don't look to make overmuch sense. Clearly, Intel has a important number of issues to resolve, and while we didn't expect them to flawlessly execute a caller level and CPU architecture release, we had hoped for a overmuch much polished product.
Depending connected nan hardware and package configuration, stableness tin beryllium an issue, though we expect this to beryllium addressed reasonably quickly. However, it's thing worthy keeping successful mind. We notified Intel astir nan mediocre gaming performance, and while they've offered a fewer solutions, nary person worked to a grade that we find satisfactory.
As it stands, gaming capacity is inconsistent, to opportunity nan least. Overall, this is very disappointing for a next-generation flagship product. If Intel can't make important improvements, Arrow Lake could extremity up being a mediocre prime for gaming.
In position of productivity, nan 285K performs better, but it's still not exceptional. While location are surely workloads wherever it excels, location are galore wherever its capacity is underwhelming compared to nan 14900K and 9950X.
One area wherever nan 285K has massively improved complete nan 14900K is powerfulness efficiency, which is awesome news. However, successful astir cases, it's still acold from wherever it needs to beryllium to genuinely compete pinch AMD.
The value is besides problematic. It's expected to beryllium astir $600+, which is acold excessively high. For nan aforesaid amount, you tin get nan 9950X aliases nan 7950X3D, and moreover nan 7950X is cheaper astatine $510. Any 1 of those AMD CPUs would beryllium a amended prime successful our opinion.
Even if Intel manages to afloat resoluteness stableness issues, hole capacity connected Windows 11 24H2, and reside nan inconsistent gaming results, nan 285K would still only warrant a value tag of astir $500. So, fixed each that, we're evidently not recommending nan Core Ultra 9 285K astatine this time, and apt won't until these issues are resolved.
As for overclocking, fixed each nan existent problems and nan request to trial capacity crossed aggregate versions of Windows, we deprioritized overclocking for this first review. It's thing we'll nary uncertainty screen successful early content, but for now, it's somewhat irrelevant fixed nan number of issues that request to beryllium addressed first.
Memory overclocking is different topic, and while nan reappraisal kits came pinch DDR5-8200 CUDIMM memory, it doesn't make overmuch consciousness owed to nan utmost value and constricted capacity gains. We're unsure of nan nonstop costs of this memory, but we cognize modular UDIMM kits are priced adjacent $300 for a 24GB kit, truthful nan CUDIMM kits will apt beryllium moreover much expensive.
Therefore, thing for illustration DDR5-7200 CL34, priced astir $115, seems to beryllium a much reasonable option. Once nan issues we've discussed are addressed, we'll look into representation scaling to find nan optimal representation support for Arrow Lake.
Wrapping this reappraisal up, we deliberation location is imaginable for Arrow Lake to get a batch amended arsenic it appears nether cooked astatine this stage. While we powerfully urge not rushing retired to bargain into Intel's caller level correct now, we're not consenting to constitute it disconnected wholly conscionable yet.
Shopping Shortcuts:
- Intel Core Ultra 9 285K connected Amazon (soon)
- Intel Core Ultra 7 265K connected Amazon (soon)
- Intel Core Ultra 5 245K connected Amazon (soon)
- AMD Ryzen 9 9950X connected Amazon
- AMD Ryzen 9 7950X3D connected Amazon
- AMD Ryzen 7 9700X connected Amazon
- AMD Ryzen 7 7800X3D connected Amazon