- Mar 3, 2017
- 1,747
- 6,598
- 136
Maybe carry a portable USB-C display with you?Laptop screens simply suck as that second desktop when I want the function of the dual monitors.
Extra 2 weeks waiting. Looks like STX has too much things to be fix, reminds me of Zen4 launch.
Maybe there's some kind of misunderstanding or miscommunication. Best Buy still has most models listed as delivery on July 15, but some models were always listed as July 28.Hyperbole much?
Besides that, if it releases on the 28th, that would mean a Sunday launch. That seems unlikely. When was the last time AMD or Intel launched a product on a Sunday?
I don't see an actual score for cb23.AMD Ryzen 9000 "Zen 5" CPUs To Offer Up To 6400 MT/s 1:1 & 8000 MT/s 1:2 4-DIMM DDR5 Memory Support, Run Cooler & Lower Power By Default
AMD's next-generation Ryzen 9000 "Zen 5" Desktop CPUs will run cooler, consume lower power & feature enhanced DDR5 OC capabilities.wccftech.com
9950x cinebench score revealed and power
The article has the tone "my cousin said he awoke and found himself in AMD's HQ where some fairies showed him the 7950X becnchmarks", so probably you don't see the score because is covered in pixie dust.I don't see an actual score for cb23.
CB23 from AT's review of the 14900KS for reference:AMD's Ryzen 9000 "Zen 5" CPUs by default will run cooler than Ryzen 7000 "Zen 4" CPUs. Based on testing with a Ryzen 9 9950X CPU (Default ES), the chip consumed up to 190W of power whereas the Ryzen 9 7950X consumes around 220-ish Watts of power in the same workloads. The workload being performed here is Cinebench R23 and we got some interesting numbers here too. The 9950X is almost as fast as the Core i9-14900KS by default and that chip can consume over 300W of power with its extreme profile.
This matches what HUB said in a podcast when he had access to a 9950X system running cinebench.Interesting details revealed there.
14900KS 12% faster than 7950X in CBR23 MT.
9950X is almost as fast as 14900KS (I guess that would be MT too).
With PBO, 9950X may "easily" (WTFtech words) sail past the 14900KS.
So this would mean that Zen 5 could be slower than Arrow Lake in CBR23 MT if Arrow Lake is 10% faster than 14900KS.
Seems like Intel has a great opportunity here to score a win, if they don't mess up Arrow Lake.
To save everyone a click:
RAM speeds are supported at 6400 MT/s at 1:1.
Can go to 9000 MT/s at 1:2 with 2 DIMMs, 8000 MT/s should be doable for most mobos with 4 DIMMs.
PBO is turned off by default, so power is not pushed out of the box. If you turn on PBO (auto-overclocking), the power will be pushed and the processor will try to clock as high as it can within power and temperature limits (95C or whatever it is).
Regarding Cinebench:
CB23 from AT's review of the 14900KS for reference:
View attachment 102052
If we assume the 9950X (Engineering Sample) can do 41000 points in CB23 MT at 190W while the 7950X does 39300 points at 220W, that's a 24% improvement in perf/W in this particular workload.
This matches what HUB said in a podcast when he had access to a 9950X system running cinebench.
The timestamp is 10:15 into the video.
I suppose we'll find out soon since we're a month away from launch. I personally think this is just a preview of what to expect but I'd be happy to be wrong.Yes, when he asked AMD about it, they removed the demo systems and later told him that the samples in the system were pre-production and were not representative of retail performance. So, assuming there is any validity to the WCCFtech story, the ES the user had could be the same issue and retail performance will be higher. That, or AMD completely lied and are trying to cover up their huge flop, lol. I think the first option is much more likely.
JF-AMD vibe?Yes, when he asked AMD about it, they removed the demo systems and later told him that the samples in the system were pre-production and were not representative of retail performance.
With how Lunar is performing in Cinebench R23 MT around 10200 on an ES at PL2 30 watts, also without HT. Arrow Lake will have faster and more P cores and 16 skymont cores. Perf/w should also improve massively because of the N3B. This will be a close match between Zen5 and LNC/Skymont client desktop.So this would mean that Zen 5 could be slower than Arrow Lake in CBR23 MT if Arrow Lake is 10% faster than 14900KS.
JF-AMD vibe?
This secrecy is getting weird.
Maybe there's some kind of misunderstanding or miscommunication. Best Buy still has most models listed as delivery on July 15, but some models were always listed as July 28.
View attachment 102038
As i see it, 7950x at 37k~39k is "almost as fast as 14900k" at 41k.To save everyone a click:
RAM speeds are supported at 6400 MT/s at 1:1.
Can go to 9000 MT/s at 1:2 with 2 DIMMs, 8000 MT/s should be doable for most mobos with 4 DIMMs.
PBO is turned off by default, so power is not pushed out of the box. If you turn on PBO (auto-overclocking), the power will be pushed and the processor will try to clock as high as it can within power and temperature limits (95C or whatever it is).
Regarding Cinebench:
CB23 from AT's review of the 14900KS for reference:
View attachment 102052
If we assume the 9950X (Engineering Sample) can do 41000 points in CB23 MT at 190W while the 7950X does 39300 points at 220W, that's a 24% improvement in perf/W in this particular workload.
Because 10 is not 32....Why am I getting a bad feeling???
Yeah, it's not a game changer like we all thought it was going to be, unfortunately. Zen 5 launching at the MSRP prices of Zen 4 is the highest they should ask for, imo, but given how cheap Zen 4 is, Zen 4 may still be the way to go. Ultimately, we should wait for official reviews before reassessing the competitive landscape.As i see it, 7950x at 37k~39k is "almost as fast as 14900k" at 41k.
So 9950x pretty much matching 14900k or beating it by say another 2K with PBO on, is not that interesting. Something closing on 50k would have been.
I just received an email from BestBuy saying that my order will arrive on 28th instead of 15th.
ASUS - Zenbook S 16