Zen 4 X3D is great - but has one Big Problem

[@garyb7193]

Excellent analysis as always! I am looking forward to 3D v-cache as another Zen 4 option we'll have this gen alongside X and non-X variances. Glad you explained trade-offs doesn't necessarily have to be a bad thing. Everything in life comes with trade-offs. It's choices and context that makes them bearable.

[@garyb7193]

@hawky2k215 yeah, but the 3D vcache is way better than I had even imagined a yr ago. It's among the best thing to happen in the cpu space in a while.

[@IGotDuhPancakeMix]

I would like to reiterate previous comments and emphasise how much I appreciate your explanation of x3d chip structure and the differences between generations. I have yet to see this information covered elsewhere. It seems you have more than earned our subscriptions.

[@oddmofo]

Really good analysis. A lot of tech channels keep talking about how AMD fixed Zen3D with Zen 4, but never stop and think about the details. I'm sure the CPUs are gonna be great for gaming, but I fear the hype and misunderstanding of the new 3D Chips are going to "disappoint" some people.

[@davidgunther8428]

The TJmax difference could also be from the public reaction to the first chips running at 95C. They've already said that they lowered the power target to 120W because of public perception of the 7950X efficiency with the 170W target. 89C looks to me suspiciously like "just under 90C". The 7950X actually has a hard temperature limit of 115C I think. You can set this in BIOS if you want to.

[@snotspat]

89C looks a lot like the second digit is the deternimating digit, so, precisely 89C, where as 90C could be "around 90", as the 9 would be the deternimating digit.

[@GholaTleilaxu]

I feel like vomiting when I read those numbers. But I guess it's all still good and "cool" as it's not 150ºC yet.

[@PineyJustice]

@@GholaTleilaxu CPU temp really doesn't even matter, it has nothing to do with heat output and everything to do with thermal transfer. 13900ks runs up to 105c and uses up to 350+w under load. The silicon can handle it just fine, the difference is high temperature packaging, the substrate and underlay glue. The high temp packaging costs more so was not normally used in desktop CPUs, only laptop CPUs. These new generation desktop CPUs are using the high temp packaging so it really shouldn't matter. As for heat output into your room/case, that's purely down to how many watts the chip consumes, watts are watts, regardless of the temp the cores are running at.

[@GholaTleilaxu]

@@PineyJustice The second law of thermodynamics matters to me, especially during hot summer days.

[@PineyJustice]

@@GholaTleilaxu The second law of shit you clearly don't understand. There is a large variable you're missing, thermal conductivity through the IHS. If the IHS was made of wood the cpu could run at 200c while putting less heat into your room than a ryzen 5800 running at 40c.

[@joaomiguelxs]

Your analysis makes perfect sense and for gamers, the trade off has been well worth it, specially on very cpu bound multi player titles. If AMD is able to keep the price to performance value at the same range as the 5800X3D, it will further consolidate its position as best value part for the segment.

[@HazzyDevil]

One point about the TDP value that I’m very iffy about. GamersNexus has mentioned that the TDP value AMD provides is based on some arbitrary formula, which they can tweak to land on a specific TDP value, and isn’t important in the real world. The formula is: TDP = tCase°C - tAmbient°C)/(HSF θca) Based off of this, they can define what the case and ambient temperature is exactly to land a number that they would like to market. Maybe something you would like to further look into? Would like to hear your thoughts on this!

[@beezle1976]

Something you've not touched on though is that the faster l3 "3d" cache with lower latencies will counter the higher reduction in clock speeds for the 7800x3d models and offer more of an increase where the clock speed reductions are lower (vs zen3 x3d).

[@knexfan0011]

One thing that could solve the thermal issues for x3D chips entirely is if they could manage to put the additional cache under the existing CCD rather than on top of it. I don't know how difficult this would be, but I see no reason as to why it would be outright impossible. This would probably require a lot of re-engineering if you start from an existing CCD design, but if the cache-under-CCD usecase were to be considered during the design of the Zen 5 CCDs it might be feasible.

[@CyberneticArgumentCreator]

7:37 not necessarily. The Tjmax is a value that means the maximum temperature the cpu allows the sensor on the die to observe before throttling/safety measures, right? But it's a heuristic based on the placement of the sensor. If the sensor is moved or if the die layout is different relative to the sensor, it will read a different temperature value than another sensor. The engineers offset this value so that it can be used as that heuristic to safely throttle and protect the die. In other words, a sensor being moved to a place that reads a slightly lower temperature at the same actual die temperature means they have to lower the tjmax value as well in order to represent the actual die temperatures at the hotspot or safe total temperature. So a lower tjmax doesn't necessarily mean earlier throttling relative to heat generated/heat dissipated.

[@Sock-qv9wr]

Very nice and concise way of explaining these new CPUs. Subbed!

[@Cakemagic1]

I've got my components for more than a month and waiting for the 3D V-Cache CPU, despite the trade off. Even if it runs slower than the normal X variants, CPU of 2022 and 2023 is already pretty blazing fast, even if it's a bit slower.

[@GholaTleilaxu]

So you are out of the 14-days or 30-days return window in case one of those components fails to work or works unacceptably.

[@Cakemagic1]

@@GholaTleilaxu That's true, but at least the company behind the components usually have very long warranties. So I'll just make use of that.

[@robojimtv]

I have seen concepts that provides microchannels within the chip designs to allow more adequate cooling of these transistors. As the technology moves towards multi layer 3D stacking, I imagine that will become a necessity to allow the chips to be well cooled.

[@KaiserTom]

A modern chip can only run at 20-30% of it's full capability due purely to thermal constraints. Rock doesn't conduct heat very well. Increasingly large parts of a die are already dedicated to structures that support heat transfer better. Modern architectures include of ton of these heat channels and more.

[@allenwebb273]

I could see the thermal issue be solved if there were some process changes that introduced the equivalent of heat spreaders in the bulk silicon, so there are less issues with hot spots.

[@Speak_Out_and_Remove_All_Doubt]

Another great video High Yield. In answering your question as to what the future of 3D stacking looking like and how will we get around the issue of stacking heat sources, the answer is on die micro channel liquid cool. To properly realise 3D chip designs TSMC have been testing Direct Water Cooling, where the water channels were etched directly into the silicon layer on top of the CPU. In there tests, silicon channels were etched into a silicon layer with a silicon-oxide thermal interface material between the microfluidic system and the actual silicon of the TTV. In a third option, the silicon-oxide TIM was replaced with a liquid metal TIM. The results were really impressive and could enable more than 2000w chips what are still properly cooled if we ever needed to these levels.

[@jeeves-2]

Excellent analysis as always, really liked the intro where you went over the rumors and if they were accurate now that we have the official specs. As for the thermal issues posed by any die stacking method, I would think if AMD is making sure the stacked cache only covers part of the die, they could use something more thermally conductive than silicon above the cores themselves to compensate for the loss in having the ihs contact the die

[@kecimalah]

I think they use silicon as it has same thermal dissipation, if they use metal it could cause some cracking etc.

[@greasebob]

@@kecimalah I wondered about this. It seems 2 small pieces of copper would be far superior to pieces of silicon, but if the pieces are bonded to the silicon below and not just "sitting" this makes sense.

[@DenisBazhenov]

Great video, thank you! The problem of power density is one of the 2 most important problems in CPU design. The other one is that price of single transistor start rising with new technological nodes. As you have said it's physics limitations. Looking at the data of power density (W/cm2) for all the CPU's from the start of 2000 you'll see that no CPU has beaten 120W/cm2. Todays processors are most dense in that sense. And the second runners are... Intel Pentium 4. Pentium 4 microarchitecture was designed for high clock speed, and therefore high power. Reaching about 100W/cm2. This also explains why M1/M2 is so much powerful and competes event with desktop chips. Apple designing their chips around performance per watt from the beginning because of limited power supply. But it was inevitable desktop becoming cooling limited. So nowadays 100W/cm2 is all you have. At least while we are waiting for GAA-transistors or some other "miracle material" to control leakage.

[@Psychx_]

As for the cache chiplet, i'd strongly assume N6. As long as you consider the maximum TSV pin density per area that N6 allows when designing the base die, stacking is not an issue. This may be an explanation for 3D chiplet boost clocks (besides core voltage limit and thermal considerations) aswell, since AFAIK the L3 also runs at core clocks. As seen with Zen 3/3+, the max possible speed for AMD's L3 implementation on a N7/N6 process is around 5Ghz, which just also happens to be the 7800X3D's max clockspeed.

[@lexzbuddy]

Does water cooling help performance much? Not AIO but how are bespoke water cooling loops performing vs air cooling and continuous boost clocks?

[@lordspam2721]

Just look at the thermals of x vs non x. The heat is why they don't want them overclocked. Great video! Very informative

[@VideogamesAsArt]

I assume the 'no clock speed regressions' was back then because leakers heard that the max boost was the same on 7950X and 7950X3D (considering same power being used for both). But that was because one of the chiplets didn't have V cache, which no-one thought about at the time

[@ThePurplePassage]

3:20 I am greatly reminded of the alleged quote that "640k RAM ought to be enough for anyone".

[@JulianDanzerHAL9001]

heat transfer isn't really affected much, it's the target temperature the cores ae designed for/set to throttle to that is lowered and thus limits how high the 3d cached die can boost

[@johndoh5182]

VERY GOOD. There are too many people who think you throw on more cache and everything is better. Well, no it's very dependent on the program AND the data, how it's accessed, if you store it, is it a data stream or just a huge set of data that keeps changing like in games, or different engineering modeling where data sets can be very large that you're actively working with. The first thing I said to myself when I saw the specs for Zen 4 is L2 is doubling. I have to think that's going to reduce the benefit of L3 because you'll have less misses in L2. To me that's a no-brainer. And for the games that didn't benefit with the 5800X3D the same will be true for Zen 4. It's once again all about the data, not the CPU. The data sets aren't changing so a CPU that has no benefit from added L3 in one generation won't get a benefit in a newer CPU either ESPECIALLY when you doubled L2. You already got all the benefit you can. Having said that many newer games will have larger data sets with more assets that a user can interact with and more data about these assets needs to be maintained as close to the cores as possible when you're close to those objects. And then there's the downside of adding cache, which usually adds latency to accessing that layer. And no, a faster CPU doesn't make you GAME any faster, it just boosts fps IF the GPU isn't a bottleneck, or already hitting 100%. So no, the data you are working with at any point in a game isn't going to increase just because the CPU is faster. It's all about the game data and a faster CPU doesn't change that if you aren't moving through the game faster, and you won't be.

[@dumbcow1]

i am so excited to get my hands on these. Finally upgrade from 7600K. My first move to team red! My CPU has been my bottleneck for many games now.

[@realsong-fake]

As a software developer I would take more cache any day with trade off of some clock speed. My reason is writing cache friendly high perf software is very costly and at the same time mental demanding. Data structure size, memory layout, cache line padding etc is just too much for most dev to the point they don't bother at all.

[@GrimpakTheMook]

7900x3d and 7950x3d will benefit quite a bit from using process lasso to schedule the threads, specially the 7950x3d

[@manicmarauder]

Sounds like they need to entirely re-evolve the thermal transport within the stacked layers instead of just relying on a massive top mounted sink to take care of it all by itself. Microscopic on die heat pipes? Electric nano-scale heat pumps? Evolve the form factor for dual sided active cooling? Less sexy but probably easier in the short term. The future will be interesting.

[@atonal440]

Vain hope: someday, we'll get to the point where amd can use 3d v-cache the other way around: put the cache die under the cpu die for better heat management. That probably requires many more TSVs though, and other changes.

[@GholaTleilaxu]

If only we would be living somewhere in space, far away from Earth's gravitational attraction...We would not survive for much time but what a happy, short life would that be, filled with 3D V-cache surrounding us on all sides!

[@JFHeroux]

My money is on AMD coming out with a much improved version of 3D V-cache in the long run. I'd bet my money on a mixed silicon and high-conductivity material for its next generation chips. Graphene could be it. I just don't know how the layer deposition would be done. But I'm sure clever engineers already have ideas on how to mass-produce such hybrid chips.

[@ArmchairMagpie]

They could re-position the cache for increased latency in exchange for lower temperatures, but that would defeat their overall design goal. Overall, when it comes to gaming, the temperatures are not much of an issue really as long as the GPU is the final bottleneck in a given situation. There are outliers which are really taxing the CPU, like Monster Hunter did at the release, but they are few and inbetween.

[@JS-wl3gi]

AM5 is just getting started, with eco modes, 3d caches, and even more cores. now we need the gpu space to get its act together.

[@VictorMistral]

What they probably could do (at a higher cost) is to put the cache under the cpu. It would mean that the silicon "padding" would need all the wiring, at probably a few electrical issues, but they could do it. And it would mean that hottest part would be closer to the cooling, it would reduce the heat issue a bit. But probably not enough to warrant the extra cost... What I don't get is why not place the additional memory of the IO die, as an L4. Sure, it would give as much of a perf increase in a single CCD, but it should solve most of the heat issue, and probably allow to clock to stay on par to their counter part without L4. So the latency would be greater to fetch from L4 then the extended L3, but still faster then system memory, and the cpu would go as fast as they can, so it would would probably be faster in some scenerio, and slower in others. (Less slow down due to clock reduction, less gain due to cache...) Add in a few DMAs, and could be used a unified cache... And could even probably be used as a "cheat" to prevent some data going through the CPU when transfering from pci-e to pci-e card, where they can't already use one of the cards DMA to prevent that... Add to that some few more features and you get a competitor to Intel DSA... If ever they decide to do some G, the L4 cache would make a bigger difference I think, as it could be used to buffer everything that goes between CPU and GPU on a high frequency (compute buffers and the likes) Intel had a line (Iris Pro) with just a little bit of eDDR as L4 (128 mb), and it worked wonders on some scenarios. Some great perf increase especially when sharing data between the iGPU and the CPU... It was eDDR, not cache, so higher latency and slower (and cheaper) I'm sure that if AMD removed 16mb of L3 from the CPU die, and added 32mb to the iodie, by default, the CPU would be close in perf. However, it would mean having a cache controller logic on the IO die, so more cost... And without full DMAs and maybe "DSA" like features (with an API can could be used by the OS) I don't think that change would be worth the additional cost. At least, until they start adding more co-processors to the CPU (hence why we do see that "System cache"/"Last Layer Cache"/L4 on ARM processors more often.

[@VictorMistral]

@El Cactuar Thanks for the correction!

[@Felix-ve9hs]

TDP = Thermal Design Power Like you said, this just tells you how strong your cooler must be

[@deus_nsf]

I know it's stupid, but I had issues with my dual CCD 3800X in the past (stuttering and single thread performance reduction) so I'm not getting anything dual CCD until it's proven mature for at least 2 generations.

[@Gottexpadder]

why did you take out the 7900x3d and the 7950x3d in the clock speed part? those two have the same boost clock as their non x3d counterpart, so your conclusion that it is a reduced clock speed fails for these two.

[@mrcrackerist]

Why wouldn't they put the CPU cores above the extra silicon layer? and could the layer under those cores be more cache?

[@WillFuI]

The 7900x3d is what I’m more excited for because I think it will gain the most from the mixed chiplet design they went with. And how this works out will determine my excitement for when AMD decides to put a high density C die and the normal computer die onto 1 package

[@JulianDanzerHAL9001]

well, it's not like the 7950x without 3d was able to boost to it's maximum boost clock on all cores simultaneously anyways

[@Pro720HyperMaster720]

Imagine if it was possible to instead of making the silicon flat on the horizontal axis and therefore the additional 3D layers horizontally stacked we could make it vertically with an IHS that cover the left and the right side, this hypothetical way would not decrease at all thermal performance on the first stack as they would be facing opposites ways that can be cooled independently, if we have two layers at the trade off of slightly reduced thermal performance we could archive the same decrease with a 100% more silicon with 2 layer on each side giving us a total of 4 layers. This technically could be even more complicated if we had a cubic 3D system where we could have 3 sides with all the cooling capabilities although with a caveat. You could have 2 vertically dies and 1 horizontal but if you want the 3 to have the same area that would create a pocket in the middle with the least thermal capabilities, I suppose it could be use for like an IO inside chip that requires a lot less cooling but is too much empty space, specially if we were to grow the size of the top and side does, as the difference between surface and internal volume differ exponentially as surface increase to the power of 2 and the volume “unusable” would grow in the power of 3. The best solution for 3D silicon I can came up with would be using the first idea of vertical silicon with layers in opposing layers AND having multiple of them in the same CPU/chip as if they were chiplets but instead of an array of horizontal “plates” we could have an array of towers of silicon all over the substrate, and if we managed to create a thermal solution that may or may not be part of the IHS that goes from the base substrate to the top of of the towers filling the gaps we could create a thermal solution that when applied to that system (if the on substrate solution we just talked about had enough thermal conductivity) would cool all silicon dies/layers equally as good or with very minor differences between the parts close to the substrate and the ones close to the cooling element on top, With this system I imagine that we could have Cubic looking CPUs or SoCs with hundreds if not thousands of layers pretty well cooled. Maybe the easiest way to visualize this would be to imagine a CPU die grow it vertically as much as it is horizontally and slice it (like with those tools that can let you slice a fruit or what ever in a single move) and in regards of how to move the data to the substrate, between two opposing layers of silicon we can have a very thin copper or something wiring

[@Pro720HyperMaster720]

Another way for the best 3D chips would be cubes of silicon designed leaving a system of microscopic pipes to use an on die water cooling, water cooling on die is actively being researched nowadays, I don’t think it was with a tube/pipeline system I think it was more like surface structures on the silicon to increase surface area for the water, but this could evolve to a pass through system that I just mention

[@NinjaForHire]

They should put the blank transistor free silicon on the bottom under and raise the good silicon.if possible I know there is a physical design issues but maybe they can do it.

[@arthurmoore9488]

You touched on it, but I wish you'd gone into more detail about the power sweet spot. As GN has shown, the X SKUs pump far more power to hit those high numbers. In performance/Watt "Eco Mode" or the non-X variants can actually beat the faster chips. The X SKUs are basically pre-overclocked for marketing reasons. Running at a lower TJMax and closer to that sweet spot just brings them down to reasonable levels. The other thing you may have forgotten to discuss was the sensitivity to higher voltages. Which is another reason Overclocking is disabled.

[@johndoh5182]

You can't really draw conclusions about heat from sensor data, unless comparing CPUs in the same generation. Sensors can be in different places, the die could possibly take more heat and a host of other issues. So while what you said is a decent idea about what happening with TDP and Tjmax, it's not fact. We will have to wait for what AMD says to know what they were dealing with in the lab. Also I'm going to disagree about the power distribution you showed for single core vs. dual core chiplet CPUs. With Zen 3, both single and dual core chiplet CPUs had the same power cap. That's not true anymore. The dual core chiplet parts have 170W TDP so each core can run hotter than previously, and dual chiplet parts can use a bit more power when running all-core loads. What you showed is basically true if NOT running all-core loads, and in fact you'd have to have the CPU running roughly 60% util. or less for that to be true. In GAMES what you showed is basically true because a 16 core part is not going to get to 50% CPU util, and the scheduler will throw threads at both core chiplets so in that case the heat is distributed better. So this is a case where you can't compete Zen 3 to Zen 4 because with Zen 4 the dual core chiplet parts are allowed to consume a lot more power.

[@hanspeter24]

thank you again for this deep dive into future technology really explaining on a die basis how silicone is gonna go the next years! 👍

[@denvera1g1]

The only reason Zen4 supports faster RAM with lower latency is because of that 12nm IO die on Zen2/3 (if you notice, Zen+ and Zen2 had similar memory limitations because the memory controllers are both on GloFo 12nm) If you use a 4000G or 5000G you can generally still beat all DDR5 kits suppoorted by Zen4 even on competitive OC boards for RaptorLake This is because the memory controller is both monolithic on dye, and on TSMC 7nm. I've got a 5700G with 75GB/s bandwidth and 49.1NS of latency, that latency is competitive to a 13900KS with DDR5 8000C36 (I'm running 4933C17-17-17-28 while in 1:1:1 timing ratio) This is just not possible on Zen4, but because Zen2 and 3 APUs have 1/2 the cache of their CPU counterparts, this results in weird performance In games where the 5800X3D accels, you're probably better off with a 5600X than with the 5700G with 4933C17 RAM But where the 5800X3D loses to the 5800X non 3D, the 5700G with fast RAM will generally be even farther ahead in performance. I'll be interested to see how fast we can get RAM on Zen4 APUs if they're ever brought to desktop, a 5nm IMC seems like it might support up to 7600C30 in 1:1 as long as you're using a 2 slot ITX motherboard and good RAM

[@anub1s15]

the thing that worries me about the 7000X3d is the thick ass heatspreader which...still not sure how / if they're gone fix that. 5800X3d ran pretty hot which I expect 7000X3d to copy but now with a thicker heatspreader to reduce the efficiency of your cooling solution. i mean I get why they did it, compatibility with AM4 coolers is nice but making the heatspreader fatter was IMO the wrong way to do it >.> it just results in more of what I absolutely hated about the early 3900X days. though there it was because it was just ramming 1.5v for god knows what reason (RGB software mostly), yea 50 degree idle on full custom watercooling is normal....what no no it's not, not while on desktop doing literally nothing. also did not realize only 1 chiplet was going to get the X3d treatment, because of how they initially presented it it seemed like a way to also overcome the issue of games running on 2 ccd's reducing performance on 2ccd cpu's as compared to single ccd cpu's I imagined it also being a way to let the 2 ccd's work more as 1...guess that didn't work out.

[@t3h51d3w1nd3r]

I wonder if in the next generation they put the 3d v cache layer underneath the cpu instead of on top, would it help the thermals a bit? Focus on cooling the cpu and letting the heat from the v cache pass through it, it might free up some performance.

[@HighYield]

I could be possible, AMD's MI300 does something similar.

[@bionicgeekgrrl]

Whilst gamers are understandably excited for x3d chips, the real market for the x3d design is the enterprise market, where applications can be built with the extra cache in mind. An example would be a f1 team modelling air flow data, the x3d chips of the milan generation were significantly faster in such specialised use cases and I expect those that can make use of it are eagerly waiting on genoa-x launching as 96 cores with x3d will be a significant increase in those applications. For desktop, it is a bit of a patch to keep pace with intel really. Until windows and other applications start being optimised to take better advantage of it, it will always be a trade off for non purely gaming desktop computers.

[@rene.s.s]

Zen 4 will have less uplift from 3D than Zen 3 did. But I do hope that the CCD handling gets improved, even over time, which might trickle down to Ryzen 9 too. I wonder if they're looking into flipping the architecture somehow, with 3D cache at the bottom. I have no idea if that's possible, but maybe there can be clever engineering gains achieved in the future.

[@DragonOfTheMortalKombat]

I don't think zen 4 will have less performance uplift than zen 3, normal zen 4 CPUs are something like 35% faster for gaming ( 7700x vs 5800x) Faster CPUs need more memory or they'll be bottlenecked. DDR5 can benefit to some extend but in the end it is also not 2x faster than something like CL16 3600 or CL17 3800Mhz that people used commonly on zen3. The best way to avoid thermal problems IMO is to design the IHS such that the metal has a small indent for V cache and the cores directly touch the IHS instead of silicon.

[@jaleelakram2232]

Bro as a techie nerd myself I love ur videos. Keep up the great work

[@winnb4968]

7800X3D is overall better gaming CPU then all 13th Gen CPU'S except ofcourse 13900K, 7900X3D is true competitor of 13900K, 7950X3D competitor of 13900KS. There is even a video for 5950X and 7950X users for a tool to use first 10 to 16 threads for games and rest for another application, this makes fps more stable especially while streaming (its tested).

[@rickyvanass7192]

I'm not an memory expert but i heard that DDR4 has a lower latency then DDR5. "Compared to DDR4, DDR5 RAM has a higher base speed, supports higher-capacity DIMM modules (also called RAM sticks), and consumes less power for the same performance specs as DDR4. However, DDR4 still holds some key advantages, like overall lower latency and better stability." So that would mean the 7000 series X3D would even more profiting from more cache then the 5000 series X3D.

[@HighYield]

At base specs yes, but DDR5 is build different internally (related to accessing the memory banks), and thus the true latency should be lower over all. But tbh, I need to dive more into this tech at some point.

[@realsong-fake]

latency means nothing without context. Let's take your typical 60 fps game or hack let's just say every game run at 240fps. Then naively speaking you would have roughly 4ms for rendering every frame. Then let's just say DDR5 is two times worse latency wise that take 100ns more for a single read. That would take up roughly more than 1/40,000 of your frame time for a SINGLE read. This is the reason why typically memory is marketed with their bandwidth per second because a single read latency does not matter (unless you are some kind of crazy use case where ns latency matters which surely not for 99.9% of users)

[@ChinchillaBONK]

Your analysis is b always the best. Easy for us layman to understand. Anyway, I feel like we make got only 5-10 years of silicon. The next company to produce next mainstream generation computers whether it's quantum computers, graphene nanotubes, sophisticated analog neural network computers , etc. Our children and children's children are gonna live in an exciting technological era.

[@jondonnelly3]

Very important argument against the 3Dvcache helping a lot in the 7900/50x3d is that the extra cache is all on one ccd and none of it on the other. The 7950x already loses to the 7700x in gaming cause of the bottleneck imposed by infinity fabric and having a lobsided cache is going to make this worse. Now the windows scheduler now has to figure out if a game process preferes 3dcache ccd or the clock speed ccd, you don't get both at the same time.

[@gucky4717]

My 5800X3D runs below TJmax and boosting to ~4,25Ghz on 105W in Cinebench. With PBO Curve Optimizer (via a thirdparty tool) and -30 on all cores, i can get below 80°C with all cores running on 4,45Ghz with 105W. I use a silent 280mm AIO for cooling. I had a 5950X before that. With all 16 cores running on 4,2Ghz, i had 65°C on all 16 cores with 142W. Or ~80-85°C with 4,5Ghz on all cores and >200W OC. That said, the hottest it got was on single core boost when it boosted upto 5,05Ghz with 1,5V (stock settings). It got to 85-95°C with just 80-85W of power. Technically both CPUs never hit TJmax in my case. It was either the voltagelimit or powerlimit. With Ryzen 7000 it is different, since they hit TJmax first above all.

[@GholaTleilaxu]

I just hope they will remember that the boiling point of water is 100ºC not 212º of the other obsolete scale.

[@Dmwntkp99]

Thick IHS is a challenge for current and 3D will be a bigger one.

[@Psychx_]

I don't think that the Windows scheduler is smart enough to only put game threads on the 3D chiplet in dual-CCD 7000X3D series. AMD can only do so much to provide hints to the operating system with CPPC i.e. by setting different "scores" for the 3D chiplet's cores. This only allows for two options however: For a new thread, the process scheduler prefers one OR the other chiplet… For max. performane I'd recommend "Process Lasso" for the time being. Btw, the logic from Intel Thread Director cannot be carried over, as it relies on extra hw registers and is a proprietary implementation. With the available information, it wouldn't be unreasonable to think that much like with Ryzen 3000, there needs to be a specific Windows update to address this and get decent scheduling decisions.

[@dex6316]

Windows 11 scheduler is absolutely smart enough. The question is whether AMD’s scheduler is, and I doubt so. Intel hybrid CPUs have a hardware scheduler that manages scheduling in conjunction with the OS. AMD appears to be relying entirely on the OS which would necessitate a Windows 11 update. AMD has historically been poor at writing code for external companies to boost performance on their products. I don’t expect the hybrid Ryzen 9 processors to perform consistently as a result. Single-thread would probably be fine though.

[@Henrik_Holst]

Most likely this will be a whitelist based solution (if people will even care as the 7950x3d surely will be a niche cpu) since this is an unsolvable problem, aka a single thread can switch between benefiting from a larger L3 cache to benefiting a higher frequency several times over it's lifespan and #1 how on earth would any scheduler be able to detect which is the case and #2 when the state changes there is a huge penalty to move the thread to the other ccd which would eat away the benefits anyway (not to mention thrashing the cache). And it's not like Intels Thread Director does an outstanding job either with it's far easier task.

[@vitalhandgrenade7210]

i mean while your spot on im not quite sure why you make a whole section to discuss clock speed and multi thread applications its already known to anyone building a Gaming PC that the x3ds are for gaming performance only.

[@CalgaryCalamari]

Wonder if they could put the 3D V-Cache below the CPU in the next version... I expect the heat generation of the cache is much lower than the CPU so having the heat-sink directly contacting the heat-sink may allow for some mitigation of the heat transfer issue...

[@leovin00]

Im curious why they can’t just put some kind of highly heat conductive metallic material instead of more silicon on top of the cores. If this was possible, wouldn’t it act like a built in heatsink and actually help with thermals?

[@kylebeyke488]

Conductivity?

[@jabcreations]

Keep up the great job on the videos!

[@Artcore103]

I haven't yet seen zen 4 benchmarks with maximum/ideal ram speeds AND timings, this will be interesting to compare with x3d. Same goes for the 5800 x3d, which was basically always compared with zen 3 parts using standard docp (xmp) settings, when we know that memory timing tuning, especially including the secondary and even tertiary timings, along with the highest achievable memory speed matched with the highest achievable infinity fabric speed, produced huge performance gains for Ryzen. I'm wondering how much of the gaming performance gap can be made up by tweaking a non x3d part this generation.

[@4dvanced112]

I’m not a chip designer so forgive my ignorance however is it not possible to have the vcache and spacer silicon in between the die and substrate instead of on top to help increase the cooling efficiency of the ihs? As in less physical silicon for the heat to transfer through.

[@alessandrosuppini943]

Hands down one of the best analysis out there of the Zen4 X3D chips by AMD, your are a champ! 👍

[@northwakes]

People say that 5800X3D scales poorly with RAM. But I like to think that it actually brings out the best DDR4 has to offer. As if you're running a 5800X with the best possible well-tuned DDR4 kit. I hope that this time around Zen 4 X3D will be able to do the same for DDR5, maybe even unleash not yet realised potential of DDR5 and be able to achieve great Intel-like performance using cheaper 5600 kits, since 7000-8000 kits that you can grab for Intel are super expensive. But of course we shall see, I'm just hopeful as a potential customer. I'm building a PC and can get 5800X3D fast, but it bugs me that 7800X3D will be potentially very good. Even if not that far from 5800X3D, it will probably increase 1% low performance quite a bit. Lows matter to me since I like to play competitively.

[@mozramdonclips]

I’ll be so excited to find out about Zen 5 3d vcache

[@somebodyrand0m280]

Thanks a lot for this video, you are very concise and informative in your analysis. My honest opinion about zen4 X3D is that I fear it is a bit too overhyped and exaggerated just like RDNA 3, which is why we wait for independent reviews. Personally, I don’t identify as a gamer as I mainly work on DAW’s and music software, which is heavily bound by single-thread performance on the CPU. And yes, I do think that Zen 3D is fantastic overall and these CPU’s may well beat the 13900KS by 10-15% in a handful of games that take advantage of that extra L3 Cache. But in other games and production workloads this won’t be the case since many apps don’t take advantage of this extra cache, which means they will mostly be slower due to their lower clock speeds. The way you explained all the caveats and trade-offs was very professionally done and as a result I learnt a lot from it. I like how you are an unbiased channel and not naively fanboying which unfortunately a lot of other tech channels and AMD fans do, which to me gets ridiculously annoying. Remember, AMD aren’t only competing against Intel, but also themselves, and it is great to see competition between the two companies. I don’t think I might need any of these CPU’s, but I am definitely looking forward to see what they are capable of when they release.

[@VADemon]

2:25 My guess is X3D's cache will see a similar performance uplift (where not GPU-bound) like for 5000-series. Why? Games are (not as much) latency-sensitive as they are not optimized to have a great memory layout IMHO. Random memory accesses = slow, such a large cache is simply so fat that even poorly laid out & accessed memory never leaves the cache. At the same time with large caches data dependencies become more of a problem, greatly multithreaded applications may spend more time waiting on those than before... careful profiling and optimization helps avoid these problems too.

[@dogu_vn7851]

Hey guys i am wondering if waiting for the new x3d models is worth it i want a all round pc iirc the 5800x3d was a beast for gaming but not so good for productivity will the same apply for the new models?

[@Psychx_]

I'd like to say that people who buy the 12-core and the 16-core X3D get a 3D chiplet and a marketing chiplet. The latter is where the official clockspeed claims stem from… The voltage on the 3D chiplet is limited to reduce thermal stress and TSV degredation btw.

[@garyb7193]

There's always gonna be tradeoffs. Intel 13900k would be even better if it had 24 performance cores instead of only 8. Yet, they market the chip as a 24-core, 32 thread design. Gotta love marketing. Gotta live with tradeoffs.

[@SweatyFeetGirl]

7950x3d is a perfect gaming chip. some games do not benefit from v-cache at all. thats where the second CCD with high clockspeeds would come in handy.

[@Pillokun]

@@garyb7193 but imagine the power draw and temps of such an 24c alder/raptor monster. :)

[@garyb7193]

@@Pillokun I image Intel would have to lower the clock quite substantially to keep power draw and temps in check. Sounds familiar?

[@Kage0No0Tenshi]

they should drop base clock and make boost clock higher

[@Hatsunari_Kamado]

Intel could fight AMD with a sepearate L3 cache section, that could solve the heat transfer.

[@maxwellsmart3156]

I think the question would be, with the thermal limitations, why didn't they put the 3D cache on the bottom? That would certainly solve much of the thermal issues but maybe the TSVs can't handle the current or the 3D cache silicon area can't support the number of TSVs required. They could use extra TSVs for heat transfer.

[@notstonks20]

The bottom of the chip is where all the connections to the interposer are

[@maxwellsmart3156]

@@notstonks20 Yes but if TSVs could be used then that wouldn't be an issue. Other than keeping connections as short as possible.

[@dex6316]

The reason why the cache isn’t on the bottom is because of packaging. There are many wires that have to be passed through from the core chiplet to the IO chiplet. These wires carry both power and data. Right now the TSVs just have to carry power for the cache chiplet and data between the cache chiplet and l3$. Flipping it would eliminate cache chiplet power delivery, in exchange for the much more power hungry core chiplet power delivery and core chiplet to IO chiplet data transfer. This will be vastly more expensive. Doing so only makes sense when designing the entire processor to be reliant on 3D stacking rather than having the 3D stacking be a bolt-on. If cost wasn’t an issue then the other trade off would be power consumption as with current technology the power would increase a bit.

[@osheaness]

Pretty sure it's the same silicon (or a different revision) on the 7nm node. 7nm has loads more capacity now that Zen 4 / RDNA 3 is on 5nm, and they PS5 has already switched to 6nm, so if they can get them for next to nothing I would imagine they would. Genoa-X might be different and could benefit from more SRAM as hyperscalers might benefit from >64MB on each die

[@snakeplissken1754]

I really would like them to release a zen4 x3d laptop cpu.... that would be amazing.

[@tuckerhiggins4336]

1:45 This is kinda wrong for the 7900x3d and 7950x3d. These cpus exhibit much less, or no clock clockspeed regressions. This is because only 1 ccd has extra v cache, so the cores that don't cab pick up the slack and clock higher Edit 9:00 - So you talked about that here. A little sloppy presentation and can be misleading, might wanna do that in one section next time

[@jaspertell6764]

I'm reminded of the dojo AI chip that Tesla designed. It is a very 3D chip development that is pushing integration into the SOC at higher level, with power delivery and heat management as well. From what I've read there is a fair amount of empty space on modern processors.. relatively. I'm a mechanical engineer that deals with thermals on a regular basis and integrating a more active cooling circuit into the chip design makes a lot of sense as you'd probably maintain or gain transistors by removing the issue that is being avoided, somewhat. Whether that technically is possible using current fabs is another story, they're pretty smart so I doubt it's been overlooked.. yet Tesla seems to be doing something about it ??

[@SanchezS1998]

HY always makes the day better.

[@nielsdaemen]

6:23 It's both! All the power you put in is converted into heat!

[@nivea878]

Intel is sinking like titanic

[@rluker5344]

I believe Zen4x3d will be faster than Zen3x3d by approximately the clock speed increase. Some of the ipc increase in Zen4 over Zen 3 is due to the increased L2. his will also help the x3d models, but to a much smaller degree. Also Zen4 was reviewed with optimal ram where Zen3 was reviewed with minimal approved standard ram. Optimal ram makes much less of a difference when you use it much less. Both of these skew the results if you are trying to estimate Zen4x3d performance as (Zen4/Zen3)*Zen3x3d. The increase in performance with clock speeds is still close to linear though so it should be safe to estimate Zen4x3d performance as (Zen4x3d.clocks/Zen3x3d.clocks)*Zen3x3d. A good analogy is from 2014 when Intel made both Haswell and Crystalwell (Haswell with 128MB L3). If you search 4790k vs 4980hq a youtube channel: "ATR Channel" will pop up. This guy did a lot of good comparison testing between the two in games that showed a lot of the same differences we see between the 5800x and 5800x3d.

[@Honigkuchenpferd187]

good vid i dont do stuff like that usually but heres my like and comment for them gains

[@JimCareyMulligan]

9:10 lower models can have two CCD

[@Pillokun]

not active.

[@MrMackievelli]

Much cheaper to use n7 for cache. The scaling beyond that is extreme hard and expensive.

[@billkillernic]

The ddr5 memory argument is bad, we are talking about a different order of magnitude comparing l3 cache to ddr5/ddr4

[@jll9764]

Great overview thanks!

[@viktortheslickster5824]

Great video, kudos for doing the research and thinking things through logically.

[@plenus7392]

Just waiting for them to decide between Intel or AMD

[@Fractal_32]

Obviously we will need to wait and see how the chips perform. With that being said personally I think the X3D chips will have a notable performance increase over their X and non-X peers. My reasoning is two fold: first the most obvious reason that AMD wouldn’t launch them if they weren’t better for (some) applications and my second reason for this belief is even with a larger L2$ a larger bank of L3$ would still be faster/lower latency then going out to main memory. (RAM) The question really is how much lower is the latency of that pool of L3$ vs calling out to main memory (RAM) and how much does that help Ryzen keep the cores fed. Additionally I would be curious if there’s performance enhancements to the IMC (integrated memory controller) which could allow X3D parts to use higher speed memory. (AMD’s IMC is currently weaker than Intel’s IMC so it’s unable to hit the same high frequency memory kit speeds.)

[@Incinerate1212]

also the price of the 7800x3d will matter alot. if its expensive.. you might as well put more money into your gpu if its gonna cross 375$.. at 4k these things are bottlenecked anyway . But i heard the true power of vcache tech is the 1% lows... it makes things smooth.. is this true?

[@19vangogh94]

Your point at 3:20 is not valid i think, L2 cache data is always a subset of L3 cache data, they are not additive in terms of amount of storage/cache.. total amount of cache is determined solely by the biggest cache number - 32MB in both examples.

[@HighYield]

I'm not sure the L3 cache is fully inclusive. We know that the L2 cache is L1 inclusive, and since L1 didnt change between Zen3 & Zen4, there is now more room inside the L2 for non L1-inclusive data. Do you have more infos or a source?

[@19vangogh94]

@@HighYield not specifically no, i just worked with software caching (i'm a software dev) so i know a bit about layered caching and usually the outer layers contain all (or most) of the inner layers data. I imagine thats the case with L1/L2/L3, bigger L2 layer cache in caching theory only means that you will access L3 less often (since you'll get more hits in L2), it does not mean that bigger amounts of data can fit into the whole caching system. Anyways it was a good video overall, gave it a thumbs up!

[@leexgx]

5900x3d and 3950x3d bit if a hogwash because of the 2 ccd been different one is higher frequency then the other and the x3d ccd has more cache but is lower clocked so windows by default will prefer the faster frequency ccd over the x3d ccd (very difficult to profile the scheduler to work like that in Windows, you see the difficulty they've got with big little cores on Intel) I would stick with 7800x3d or 5800x3d if your still on a 3000 or older cpu (2000/1000 with a simple bios update gives a lot of performance uplift)

[@cosaovidiu2838]

Excellent video and analysis!

[@knofi7052]

Great analysis! However, I think, it's still a good solution! And there will be more to come with this technology...

[@BlueRice]

isnt the point of 3D was mostly for gaming? higher clock speed doesnt always mean faster gaming.

[@mjes911]

Great video! Keep up the great work

[@jondonnelly3]

Fantastic overview. Well done

[@danclifford9462]

Your a little out of date. AMD have already stated that teh 3D Cache CCD on 790000 and 7950 are limited to 5Ghz while the non 3D CCD has standard boost

[@ceuser3555]

Very informative, thank you.

[@markosimic6305]

R u from Germany?