Sponsor: Get 10% off Squarespace purchases (
This deep-dive looks at the AMD Zen 5 CPU efficiency for the Ryzen 7 9700X and Ryzen 5 9600X, introducing gaming efficiency and compression efficiency to our testing suite. The benchmarks find that AMD’s Zen 5 is at times extremely efficient, like in some all-core workstation tasks (such as rendering), but can also be remarkably inefficient in some gaming scenarios. Even when FPS-normalized (locked to 60FPS) or power normalized (locked to 50W PPT), the Zen 5 CPUs can prove less efficient than Zen 4 in many scenarios. This also introduces testing of the 7700 non-X and 7600 non-X, as those were lower TDP parts more directly comparable to the 9700X and 9600X for power testing.
The best way to support our work is through our store:
– use code THISISFINE for 10% off while it is active.
Like our content? Please consider becoming our Patron to support us:
RELATED PRODUCTS [Affiliate Links]
AMD R7 9700X on Amazon:
AMD R7 9700X on Newegg:
AMD Ryzen 7 7800X3D on Newegg:
AMD Ryzen 7 7800X3D on Amazon:
AMD R7 5700X3D on Amazon:
AMD R7 5700X3D on Newegg:
TIMESTAMPS
00:00 – Zen 5 Power Efficiency Research
03:18 – Don’t Buy Zen 5
05:51 – Testing Methodology
07:45 – Definitions & Efficiency
09:39 – Blender Power Efficiency
10:51 – Cyberpunk Phantom Liberty Power Efficiency
12:35 – 10-pt & 30-pt Highs in Cyberpunk
13:50 – 60FPS Locked Power Comparison
14:21 – Cyberpunk CPU Voltage
15:35 – FFXIV Dawntrail CPU Efficiency
17:27 – FFXIV Dawntrail Power vs Time
18:39 – FFXIV Max VID
19:15 – FFXIV VDDCR SVI3
19:50 – FFXIV All-Core Frequency
20:19 – Stellaris Efficiency
21:45 – F1 24 Benchmarks
24:37 – Starfield Efficiency
25:55 – Baldur’s Gate 3 CPU Benchmarks
26:47 – Rainbow Six Siege CPU Benchmarks
28:12 – 7-Zip Compression CPU Benchmarks
29:25 – Stellaris Highs
30:14 – FFXIV Dawntrail Highs
31:03 – Electricity Cost Per Year
32:53 – Conclusion
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video (“this video is brought to you by”) and above the fold in the description. We do not ever produce paid content or “sponsored content” (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
t:
f:
w:
Steve Burke: Test Lead, Writing, Host
Patrick Lathan: Testing
Mike Gaglione: Testing, Video Editing
Jeremy Clayton: Testing, QC
Vitalii Makhnovets: Video Editing
[ad_2]
source
Help support our testing! New colorful solder mats on the store, plus copper plated mule mugs in stock! Use code THISISFINE for 10% off while it is active: https://store.gamersnexus.net/
Check out the latest hardware news here: https://www.youtube.com/watch?v=ZRJaF2ky9CE
According to MLID Windows was not ready when they launched ZEN 5. They had an open goal to put the game away but they hit Intel's side post! I think reviewers would need to do a reboot when they can get Windows to work with ZEN 5.
Huge thanks for this incredibly detailed benchmarking and ESPECIALLY for including the 5600/x. The chart of enery cost per year for each CPU was also extremely awesome (i'm one of those people at 40cents/KWh…).
I would just like to add a little bit of perspective to my personal situation and view on hardware: The main reason i care about power consumption is heat. I do not have air conditioning in my gaming room (because my electricity is extremely expensive) so i try to have my gaming PC produce as little heat as possible, which USUALLY means consuming as little power as possible. Saving up to 40 bucks a year on the power consumption of the CPU is "nice" – but not a make it or break it argument – installing and running AC would cost me multiples of that amount. But since my gaming room is not ACed, i DO care a LOT about the heat my PC is pumping into it. Right now, i'm sitting at 28Β°C and sweatting. It's not comfortable at all. My PC adding 6Β°C instead of 2Β°C really matters. I can "feel" that difference and it impacts me personally more than a 20 or even 30% difference in performance would. But just minimizing the absolute power consumption to reduce heat is not the solution, either, as i still want to actually game. Looking at 15 fps isn't fun, at that point it would be better to just do something else. So efficiency IS important to me – but the absolute power consumption is, too.
This video did not look at temperatures at all and GN always does their thermal testing at 21Β°C ambient – but it DID include the absolute power consumtion in all of its charts, which is almost as good.
I'm currently running a 5600X and exclusively play at 1080p – so the CPU is actually important to me. My conclusion from this video is that the 7800X3D is basically identical to my current CPU in power consumtion – it's just a lot faster. I would really like to see/hear something about the 7800X3D limited to 50 watts as it wasn't that much higher in most charts, anyway. How much performance are you losing by limiting the 7800X3D to 50 watts? Does it get noticeably cooler? because i have seen reports of the 7800X3D getting quite a bit hotter than the 7600 or 5600X.
AMD and Intel have fully switched roles now
After religiously following GN, it feels like no other tech outlet can give me better reviews on hardwares
That was a lot of work to put in. I respect you for thinking about how you can address efficiency.
Re Zen 5 this is the strangest CPU launch I can remember partly because in the Zen era we're used to AMD delivering the promised performance give or take a little bit. But this launch has had the recall of first batch CPUs, a disconnect between the marketing of the CPU vs actual performance/what's in the reviewer's guide (per Hardware Unboxed), core parking and admin account (Jay), and now you going at the efficiency topic to show that the reduction in TDP isn't necessarily the same as efficiency. I don't think I've seen so much odd stuff happen in such a short space of time for one CPU generation. And none of them are particularly bad individually, but collectively it feels like a mess. But as Jeremy Clarkson said, 'Still, could be worse': if ever there was an ideal time for a mess to happen it's now with what's happening at Intel.
Is this AMD generation about getting architectural changes in place for the future and then building on top of them with the next couple of generations?
awesome video
No reason to get 9series, 5 and 7 seems fine. I'm just going to wait for am6
"Loading screen simulator!" πππ π π π
It is clear that some of the problems will be solved over time, and after six months/year the price will drop, but yes – 9000 is a light disappointment
Intel dodged a bullet….
Personaly i'm most intrersted in the Idle / Low load scenario. What happens if i leave the PC runnung 24/7.
Also testing the efficancy in general shows, if they have actually developed a better cpu or just buying more performance with power consumtion.
What CPU to fulfill the Paris agreement and stay within 1.7Β°C of global warming ? π
You can literally design a render workload to stress parts of a CPU differently, Scenes heavy on procedural textures but light on image-based textures stress differently from extremely geometry heavy scenes or scenes with huge 8k PBR textures slathered all over the scene. I've literally done this on purpose while designing benchmarks.
i feel like the 4090 connector issue is getting a pass. they are still melting all around the world
I am still very satisfied with my 7950x in 105 eco mode in my lianli a4 h2o mini beast π
The latest AMD chips are not necesarily the most efficient? Speechless.
Btw, things I like the most is the data at 31:05. Both are practical and helpful for a layman like me.π
Thanks Steve and the team!
12100f gang stays winning I guess
I don't understand this game centric FPS per watt sillynous. Can't you be different and do a single financial optimization benchmark in Windows? Games change every year, MonteCarlo simulation remains the same. Threadripper is not always the solution for this kind of work because… The next mistake you make is running the tests in a large building that can absorb all the computer heat. Try running an i9-14900k or TR in your bedroom 24/7 in Phoenix. It's not the cost of electricity that matters, it's the cost of air conditioning!
I'm glad the reviews expanded. If you add 10 minutes, I'll watch it anyway; it's not like I watch any video in 1x. I'll bump the speed by 0.1 or 0.2 if it gets too long. I don't think I'm unique among this audience, so longer videos aren't necessarily bad. For people who struggle with speed-up video, get an addon that lets you have more granular control and slowly build up.
So zen5 is not good?
Is AMD scamming people for selling 7k series branded as 9k?π
Jesus Iβd kill for electric that cheap! $0.10 kWh is cheaper than what we get here in British Columbia and we literally sell electric to the US. We generate electricity and sell to all over the western US and they get our electricity cheaper than we pay! Itβs criminal. Iβm paying $0.1416 kWh