The Intel Core Ultra 7 265K features 8 P-Cores and 12 E-Cores with improved clock speeds and power efficiency. Let's see how it compares to AMD's lineup and the i7-14700K it means to replace.
Intel Core Ultra 7 265K Review: Better Value, But...
- Thread starter Steve
- Start date
MarcusNumb
Posts: 217 +378
It's funny that I feel so hyped up about the 9800x3d when I read reviews about Intel CPUs like this. Anw, just control the expectations since AMD loves to kick their own a**** xD
I can't see why AMD would release the 9800x3d if it's only competing with itself? Wouldn't it be better to wait for Intel to catch up then release an x3d model that beats their latest offering.
ScottSoapbox
Posts: 1,355 +2,445
Because they already spent the time and money to create the 9800x3D. Now, they need to sell it to fund the development of Zen 6 and Zen 7.
Companies can play with the release timing a bit, but the only way they have the resources to create the next chip is by selling the current chip. Waiting until Intel's next release would hurt AMD's lead in chip development.
yRaz
Posts: 7,115 +11,051
The chip has already been in production and taking up warehouse space for several months now. People don't understand that most of this stuff is financed with debt for tax reasons but they can only hold onto it for so long before the interest on the debt exceeds the amount of money saved on taxes.
Also, the question itself doesn't seem to understand that businesses who stay stagnant and don't innovate usually fail when they become comfortable in their market position.
People also don't seem to understand that these products take can decades of development. It isn't that zen 1 is out, let's start working on zen 2. They were already working on zen 3 while they were finalizing zen 1. These products take 6-7 years of development and lots of time working with manufactures to see if anyone can actually make the product that is being designed. This literally just happened with Nvidia and TSMC where TSMC had to use their knowledge of manufacturing to improve nVidias design to increase yields.
Too many people have Brain Rot from TikTok and YouTube shorts.
Last edited:
PierreJG
Posts: 35 +9
Intel did the same and Nvidia still does it. Where Intel jumped in front I think was in 2008-2009, Intel released I think 8 cpu's in 1 quarter. Then amd Brought out CMT which we all know was a massive fail. Intel released CPU's every year even though AMD was so far behind, Now after 10 years, AMD reversed this.
It is not fun as competition is what makes that market what it is but no company will wait for another company to catch up as it goes about profit. Why buy a weaker cpu when you can get the fastest cpu for around the same price or a little more.
VemigoAmigo
Posts: 130 +89
Because 7800X3D is not perfect at all. Application performance is poor due to gimped clockspeeds. This is what 2nd gen 3D cache is going to fix, by placing the cache below the CCD instead of on top = Less heat, less throttling = Higher clockspeeds.
9800X3D has like 20% increased ST perf and 30% increased MT perf stock vs stock + option to run even faster due to multiplier being unlocked. OC is possible. 3D cache is way less sensitive this time.
For people that likes a balanced system - people doing both applications and games - vastly higher multithreaded performance will happen with 9800X3D.
9800X3D beats 9700X in applications in leaks. Now the name actually makes sense.
9800X3D > 9700X.
In comparison, 7800X3D loses to 7700X, sometimes even 7700 NON-X, in applications. Name did not really make sense. It beat 7700 series only in games.
9800X3D will get around 10% gaming uplift (average, far cry 6 shows 13-16% vs 7800X3D) and a 20-25% application uplift which is very good to see. Again average, Cinebench MT shows 30% uplift.
I see 9800X3D as a somewhat perfected 7800X3D. Solves the limitations of 5000X3D and 7000X3D and increases clockspeeds.
The low clockspeed of 7800X3D has been a problem in some games as well. Look at Starfield and you will see that i5-13600K beats 7800X3D for example. This won't happen with 9800X3D.
Many games likes more cache but some still want high clockspeed and 9800X3D is going to deliver both.
Last edited:
In most gaming load, this chip can barely distant itself from the i7 12700K which may cost half its price and not that different in terms of power efficiency. In other productivity workload, the Arrow Lake chip may be better, but I can’t recommend it for pure gaming machines. So I think we can expect worst performance on laptops when the SODIMM are still far from fast.
VemigoAmigo
Posts: 130 +89
Most chips don't distant themself from a i7-12700K in 1080p tho:
https://www.techpowerup.com/review/intel-core-ultra-7-265k/19.html
And most people today uses 1440p or higher I guess.
By most I mean people that actually cares for higher end hardware.
If you want the best CPU for gaming, just wait for 9800X3D.
phinn888
Posts: 22 +18
Was really hoping Intel would make a comeback, cutting edge TSMC 3N lithography and new architecture and we get THIS? It's not even that power efficient or good at productivty just look at Photoshop, and gaming is about 2 generations behind AMD now.
What happened Intel? My next build will likely now be a 9800X3D + B850 mATX.
What happened Intel? My next build will likely now be a 9800X3D + B850 mATX.
From all the leaks, gaming performance will seemingly be the worst increase, which will of course still easily make it the best gaming chip on the market. The productivity performance though will be the real story as it will match or exceed the 9700X, unlike previous X3D which were behind their non-X3D counterparts in productivity. But if you are expecting a 25% increase to gaming performance, from what has leaked that seems unlikely, probably around 8%.
ScottSoapbox
Posts: 1,355 +2,445
True on the warehoused chips, I was simplifying a bit for brevity.
Relatedly, the development lead time is also why companies can appear slow to correct issues because if an unknown issue slips through, it can't always be fixed on the next-gen that is already too far along, so it's two generations to fix it.
yRaz
Posts: 7,115 +11,051
I was more replying to than the person you replied to than just you. But I had this thought the other day after making this post. It's like people think that computer parts are like making fast food. You order an AMD burger through Amazon, Amazon sends AMD the order. AMD cooks it up for you and in a few days you get it and that's just not how it happens.
People need to learn product date codes. Once you know where to and how to read them the world is a different place.
ScottSoapbox
Posts: 1,355 +2,445
LOL. I don't think it's quite fast food speed, but I agree that many people act like it's a few weeks maybe a few months from idea to finished product. Those people also believe new cars come out every year.
It's like people think companies haven't figured out the GPU could be faster with more cores. After this annual epiphany for "more cores," they turn some knobs on the big machines and start cranking out products for testing.
yRaz
Posts: 7,115 +11,051
jpvalverde85
Posts: 43 +29
I came to see a spaking, and I got served well! Not only meh, is goes backwards. It may be good at mobile, but at desktop this arch mostly never manages to stretch his legs.
whateversa
Posts: 194 +231
Why do articles still talk about ghz speeds? It is not a reflection of actual performance.
Because clock speed can be actually measured. "Actual performance" depends on gazillion things.
OK - so gaming this is not great - but neither is it bad. You are still getting > 160fps. Where it DOES shine is in productivity cases such as this is ideal for a Plex server. You can TRANSCODE 16 streams 4K video without any issues on the iGPU and total system load was around 60 watts (and it also has native transcoding for AV1 as a bonus!). So I can build a low wattage plex server vs an AMD server that requires a Nvidia GPU of 4070 or greater, adding another ~300 watts to the picture and another $500-$600. You also have to contend with added heat. Added bonus is the additional NPU (although I still am not a believer of AI - which is still not truly AI). Sure, the PCIe # of lanes suck compared to AMD for PCIe5 and 4. Also, Intel and AMD have not standardized a lot of instruction sets (although they are now working together to bridge this gap), so the safer bet is with Intel.
Similar threads
- Locked
- Locked
Latest posts
-
The OLED Burn-In Test: 9-Month Update- ddferrari replied
-
Company takes pre-orders for AI workstations with eight RTX 5090 GPUs, priced over $52,000- StupidPeopleSuk replied
