Review Zen4 3D review thread

[Hitman928]

Thread to focus on Zen4 3D cache CPUs.

New gaming king (as most expected), though the 2 CCD 7950x3d does seem to have issues with some games, more than I would expect of it getting stuck on the "wrong" CCD. I imagine it will get cleared up with subsequent updates but we'll see. Simulated 7800X3D showed no such issues and overall has the gaming lead (real product might be slightly slower though depending on in game clocks).

Computerbase also has the 7950x3d as the gaming champ. They (and TPU) also show that efficiency while gaming is extremely good.

Ryzen 9 7950X3D im Test gegen 5800X3D und 13900K(S)

Zen 4 3D ist da. Im Test erweist sich AMDs Ryzen 9 7950X3D gegen 7950X, 5800X3D und Core i9-13900K(S) als schnell und effizient.

www.computerbase.de

Just to toot my own horn a little, it landed spot on with my prediction of fastest gaming CPU but not significantly so over a 13900k on average, but with much higher efficiency.

Additional reviews, will add more later.

AMD Ryzen 9 7950X3D processor review

We review the new Ryzen 9 7950X3D processor from AMD. The much-anticipated processor series comes with an added cache that will help predominantly with gaming. This processor series was designed for ...

www.guru3d.com

AMD Ryzen 9 7950X3D Review - Best of Both Worlds

The Ryzen 9 7950X3D is the spearhead of the AMD Zen 4 X3D lineup. In our performance review we test AMD's new 16-core flagship with dozens of applications and 14 games at up to 4K, to see whether AMD can take the performance crown from Intel, how power consumption is affected, and what can be...

www.techpowerup.com

The AMD Ryzen 9 7950X3D Review: AMD's Fastest Gaming Processor

www.anandtech.com

Gamers Nexus

AMD Ryzen 9 7950X3D Linux Performance Review - Phoronix

www.phoronix.com

AMD Ryzen 9 7950X3D Review: AMD Retakes Gaming Crown with 3D V-Cache

AMD takes a bite out of Raptor Lake.

www.tomshardware.com

AMD Ryzen 9 7950X3D review: Team Red retakes the lead with its best CPU ever

The AMD Ryzen 9 7950X3D is the best gaming CPU there is, and it's not close

www.techradar.com

 

[biostud]

Electricity everywhere is usually priced at an hourly rate. The idea is that when usage increases, the price should also go up to incentivize people to cut back or to delay consumption until off-peak hours.

Here you can usually buy, with hour rates, quarterly or yearly fixed rates. But it is quite of topic from this thread.

 

[In2Photos]

My bill
Energy charges 685.42 Basic Charge 11.00 Energy Use Charge (1,000.000 kWh x $0.06642) 66.42 Energy Use Charge (4,202.000 kWh x $0.07002) 294.22 Transmission Charge (5,202.000 kWh x $0.00585) 30.43 Distribution Charge (5,202.000 kWh x $0.05447) 283.35 Regulatory charges and credits 81.82 102 RPA Exchange Credit (5,202.000 kWh x $-0.00676) 35.17 (CR) 105 Regulatory Adjustments (5,202.000 kWh x $0.00016) 0.83 109 Energy Efficiency Funding Adj (5,202.000 kWh x $0.00579) 30.12 110 Energy Efficiency Customer Svc (5,202.000 kWh x $0.00004) 0.21 112 Customer Engagement Transformation Adjustment (2,511.310 kWh x $0.0003) 0.75 118 Bill Adjustment Cost Recovery 1.14 123 Decoupling Adjustment (5,202.000 kWh x $-0.00222) 11.55 (CR) 125 Annual Power Cost Update (5,202.000 kWh x $0.01007) 52.38 126 Power Cost Variance Mechanism (5,202.000 kWh x $0.0008) 4.16 135 Demand Response (5,202.000 kWh x $0.00094) 4.89 136 Community Solar Cost Recovery (5,202.000 kWh x $0.00021) 1.09 137 Solar Payment Option Cost Recov (5,202.000 kWh x $0.00005) 0.26 138 Energy Storage Cost Recovery (5,202.000 kWh x $0.00012) 0.62

Hard to read, but its essentially $0.15 per kwh. And Europe appears to be about $0.52 or more than 3 times our rate. So the difference between a 142 watt 7950x and a ~250 watt 13900k is considerable per year. My 64 core 7742's are only 225 watt !

Your bill isn't that hard there's just a lot of different charges. Essentially you pay a basic facilities charge, an energy charge for the first 1,000 kWh and a different rate for everything above the first 1,000. Then you pay a bunch of different charges or receive credits for various things.

 

[Markfw]

Your bill isn't that hard there's just a lot of different charges. Essentially you pay a basic facilities charge, an energy charge for the first 1,000 kWh and a different rate for everything above the first 1,000. Then you pay a bunch of different charges or receive credits for various things.

Point being.... Those not in the USA, definitely Europe, pay more than 3 times what we do, and the gaming wattage can make a significant difference in your power bill. Just one more reason to dump your 13900k, maybe even the 13700k or even the 13600k (would have to evaluate those), in favor of a 7950x3d.

And for DC work, those in Europe need to be rich to do it, as my usage @ 0.52 per kwh = $2,707 !!!

 

[Mopetar]

I'd honestly wait for the 7800X3D. You'd have to game for a lot of hours to make up the difference between a 13600K and a 7950X3D to justify the price difference, particularly when we know that Intel isn't nearly as inefficient in gaming loads where it's not trying to run every core well past the efficiency curve.

If you already have an Intel CPU it's probably not worth changing unless you can get at least 90% of what you paid not just for the CPU but the board as well. Even if we just use raw TDP values and the worst possible case for Intel and the more expensive European power on top of that, it's a nickel per hour more expensive for an Intel user.

Never mind that in most titles the CPU difference won't show up unless you've got a top-end GPU, and that's going to be sucking down more than the CPU. Last generation a 3090 was almost an order of magnitude more than a 5800X3D and while it's not quite as bad this time around, top-end GPUs are still going to make the CPU look a lot less significant by comparison.

 

[biostud]

I'd honestly wait for the 7800X3D. You'd have to game for a lot of hours to make up the difference between a 13600K and a 7950X3D to justify the price difference, particularly when we know that Intel isn't nearly as inefficient in gaming loads where it's not trying to run every core well past the efficiency curve.

If you already have an Intel CPU it's probably not worth changing unless you can get at least 90% of what you paid not just for the CPU but the board as well. Even if we just use raw TDP values and the worst possible case for Intel and the more expensive European power on top of that, it's a nickel per hour more expensive for an Intel user.

Never mind that in most titles the CPU difference won't show up unless you've got a top-end GPU, and that's going to be sucking down more than the CPU. Last generation a 3090 was almost an order of magnitude more than a 5800X3D and while it's not quite as bad this time around, top-end GPUs are still going to make the CPU look a lot less significant by comparison.

And for the same reason it is a good idea to explore the possibility of undervolting your GPU, as you probably can save a lot of electricity and only loosing little performance.

 

[DAPUNISHER]

I think the energy use discussion is missing caveats. Tech Testers showed the 7950X 3D idling at 40-50W on the Asus board they used. The firmware needs to mature, because that is unacceptable. Which makes power calculations tough, as you can't count on your AM5 board to use similar power to another. The result is throwing around numbers for power usage is highly speculative.

E.G. I spend a lot of time online reading and watching content. 50W idle would be 5 to 6 times the idle of my 5800X = that board would be going back.

 

[Mopetar]

And for the same reason it is a good idea to explore the possibility of undervolting your GPU, as you probably can save a lot of electricity and only loosing little performance.

Aren't a lot of the people who are undervolt doing so in order to get higher clock speeds? The overall power is still similar, you just get more from it.

Though with current GPUs you can usually drop the power limit by a fair bit and still get around 95% of the performance.

 

[In2Photos]

I think the energy use discussion is missing caveats. Tech Testers showed the 7950X 3D idling at 40-50W on the Asus board they used. The firmware needs to mature, because that is unacceptable. Which makes power calculations tough, as you can't count on your AM5 board to use similar power to another. The result is throwing around numbers for power usage is highly speculative.

E.G. I spend a lot of time online reading and watching content. 50W idle would be 5 to 6 times the idle of my 5800X = that board would be going back.

It's not just the 7950X3D. My R7 7700 idles at almost 40W with EXPO enabled. I used a Kill-A-Watt to measure my PC at load and idle. 160w @ idle, 416W during a gaming benchmark. My daughter's 7600X uses about 10W less than my 7700 @ idle. I would love to see that improve with firmware updates.

 

[Hail The Brain Slug]

I think the energy use discussion is missing caveats. Tech Testers showed the 7950X 3D idling at 40-50W on the Asus board they used. The firmware needs to mature, because that is unacceptable. Which makes power calculations tough, as you can't count on your AM5 board to use similar power to another. The result is throwing around numbers for power usage is highly speculative.

E.G. I spend a lot of time online reading and watching content. 50W idle would be 5 to 6 times the idle of my 5800X = that board would be going back.

Your 5800X must be running at 3200mhz or lower ram speeds to have that low of an idle. Any ram overclocking above the rated memory frequency of the CPU causes the SOC to run in "overclock/full power" mode all the time, amplifying idle power consumption.

Granted that, my 7950X3D idles at around 25W PPT even with EXPO + ram overclock, which is around what my 5950X idled at with very fast ram.

 

[DAPUNISHER]

Your 5800X must be running at 3200mhz or lower ram speeds to have that low of an idle. Any ram overclocking above the rated memory frequency of the CPU causes the SOC to run in "overclock/full power" mode all the time, amplifying idle power consumption.

Granted that, my 7950X3D idles at around 25W PPT even with EXPO + ram overclock, which is around what my 5950X idled at with very fast ram.

Pardon me for being a fool and a lout. I was looking at CPU+SoC, my PPT is only a couple of watts lower than yours'. I run 4x8GB 3600 XMP.

 

[Kocicak]

Talking about energy efficiency in gaming, I just tested what the CPU does while I am running around the main square of Hogsmeade, and the CPU shows cca 50% utilisation on 6 P cores (main threads) and pulls around 50W. Both CPU and GPU 6750XT show nearly zero power draw in idle.

 

[igor_kavinski]

Is the game the only process consuming more than 16GB???

 

[Kocicak]

This game alone consumes 9-11 GB of RAM, I tested more locations now.

Edit: now the game decided to use half of that. What is happening. Perhaps before it used even more than 11.

 

[APU_Fusion]

Clearly, the 13600k is superior. https://cpu.userbenchmark.com/Compare/Intel-Core-i7-4790K-vs-AMD-Ryzen-9-7950X3D/2384vsm2052977

😂

 

[Just Benching]

Talking about energy efficiency in gaming, I just tested what the CPU does while I am running around the main square of Hogsmeade, and the CPU shows cca 50% utilisation on 6 P cores (main threads) and pulls around 50W. Both CPU and GPU 6750XT show nearly zero power draw in idle.

View attachment 78103

All Intel cpus consume the same power in hogwarts. THe game doesn't really utilize the CPU - just a couple of cores.

 

[Just Benching]

Clearly, the 13600k is superior. https://cpu.userbenchmark.com/Compare/Intel-Core-i7-4790K-vs-AMD-Ryzen-9-7950X3D/2384vsm2052977

😂

The 7950x 3d is a good cpu - but the 7800x 3d will be much better and easier to work with out of the box. The problem - tradintionaly - with amd cpus for games, is what is now called "the amd DIP". Haven't checked a lot of games, but warzone 2 that works great on AMD hardware - 1% lows on the 3d are just bad , worse than my stock 12900k. Averages are of course much higher, but 1% lows are horrific.

 

[inf64]

Clearly, the 13600k is superior. https://cpu.userbenchmark.com/Compare/Intel-Core-i7-4790K-vs-AMD-Ryzen-9-7950X3D/2384vsm2052977

😂

What a disconnect from reality. The whole site is spammed by bot comments and fake user rating, it's so obvious.

 

[Markfw]

The 7950x 3d is a good cpu - but the 7800x 3d will be much better and easier to work with out of the box. The problem - tradintionaly - with amd cpus for games, is what is now called "the amd DIP". Haven't checked a lot of games, but warzone 2 that works great on AMD hardware - 1% lows on the 3d are just bad , worse than my stock 12900k. Averages are of course much higher, but 1% lows are horrific.

Saying that lows are horrific, based on this is just trolling. Its 20 fps higher on 1% lows from the 12900.

 

[coercitiv]

The problem - tradintionaly - with amd cpus for games, is what is now called "the amd DIP". Haven't checked a lot of games, but warzone 2

Ladies and gentlemen, the "the amd DIP":

 

[Markfw]

Ladies and gentlemen, the "the amd DIP":
View attachment 78167

So even though the 7950x3d is the best in overall 1% lows, you found one game where its not the best ? only second and its only 6% slower on this one game ? That is called "the amd dip" ?

 

[Shivansps]

This game alone consumes 9-11 GB of RAM, I tested more locations now.

Edit: now the game decided to use half of that. What is happening. Perhaps before it used even more than 11.

Did you disabled virtual memory? dont do that, Windows needs the pagefile to assign unused memory requests.

 

[inf64]

So even though the 7950x3d is the best in overall 1% lows, you found one game where its not the best ? only second and its only 6% slower on this one game ? That is called "the amd dip" ?

I think that video shows how superior x3d is

 

[Just Benching]

Ladies and gentlemen, the "the amd DIP":
View attachment 78167

Yeah, that's exactly my point. Look at the 1% lows
It's worse than a stock 12900k. A tuned 3d gets around 150 to 160 in quads matches, a stock 12900k gets 170-180

 

[Markfw]

Yeah, that's exactly my point. Look at the 1% lows
It's worse than a stock 12900k. A tuned 3d gets around 150 to 160 in quads matches, a stock 12900k gets 170-180

12900k is not even on that list. So even though its average is the best of all cards in 1% lows, in ONE game, ONE cpu is better and you call that horrific. That is trolling.

 

[Just Benching]

12900k is not even on that list. So even though its average is the best of all cards in 1% lows, in ONE game, ONE cpu is better and you call that horrific. That is trolling.

I wasn't talking about the list but about the video he posted.

It's not the only game, now I'm testing cyberpunk.

It is horrific in the sense it's losing to a stock 12900k - that cpu is slower than a 13600k man...