Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AMD Ryzen 3000 announced

anandtech.com

Post reply on HN
Print view View original

341–347 of 347 posts

Re: AMD Ryzen 3000 announced

#341

Post by TheOperator »

zelon88 wrote: post #106
TheOperator wrote: post #56

Earlier quoted context omitted.

Also with CPU performance starting to taper off I think people need to also start think of their computers as longer term investments. I think the 8 core makes sense. Even my phone has 8 cores now...

I've been hearing this for a long time, but the "taper" isn't really a taper in all markets. In consumer markets, yes the taper is very real. Infact I bet you could take one $500 machine off the shelf of Wal-Mart every year for the past 5 years and benchmark them all within 15% of each other overall. But it isn't really visible when money is no object. If it's not the cores getting faster it's the core-count going up…

That's just it. The 12 core system years from now will have barely aged. CPUs will just have gotten larger and more expensive.

This is changing a past situation where say a CPU would lose half its value in 3 years. This incentivizes spending big and upgrading infrequently.

Re: AMD Ryzen 3000 announced

#342

Post by teknologist »

gigatexal wrote: post #150
tmd83 wrote: post #105

I'm really looking forward to this. But one issue with current AMD cpu is that you have to buy a desktop GPU even when you are not doing any kind of gaming and such. I know intel iGPU haven't been terribly good but for work they are good enough and is one less part and cheaper to boot. For same performance a Ryzen 3rd gen + gpu might still be cheaper but the price advantage gets reduced. I haven't really seen that me…

The G-line of CPUs with integrated graphics is also interesting to me as well! As system ram gets quicker (it's my understanding that the frame buffer for these integrated GPUs comes from system ram) and nanometer tech gets better that a competent integrated GPU+CPU chip will be able to play AAA games at 1080P with some of the visuals turned up. Though the conspiracy theorist in me thinks that the console makers woul…

Since console manufacturers seem to have been favouring AMD for graphics lately, this could actually be the reason for a lack of AMD iGPUs

Re: AMD Ryzen 3000 announced

#343

Post by nolok »

effie wrote: post #311
XMPPwocky wrote: post #259

Earlier quoted context omitted.

Ah, so AMD doesn't have cheesy numbers- good to know.

Not entirely true, some AMD processors too can draw more power than the TDP value would suggest, due to XFR.

Agreed but the effect is much more limited. If you compare manufacturer TDP against each other, it can only go even more in AMD's favor in real world scenario.

Not very surprising even only from the manufacturing process used difference though, since Intel's next desktop generation is still 14nm it won't be changing soon.

Re: AMD Ryzen 3000 announced

#344

Post by nolok »

fencepost wrote: post #59

My question is how is their architecture doing regarding the assorted speculative execution baked-in issues and what kind of impact is there on the AMD processors compared to comparable Intel CPUs?

Those affect Intel much, much more than AMD and once you include the fixes it always get better for AMD than without because, oversimplifying a lot, when doing branch speculation "Is it branch A or B ? I will guess A and pre compute that" Intel didn't bother with any access check during the prediction and only ran them after once confirmed it was indeed A (so it already had the result and merely did the ACL check), whereas AMD was already doing all the checks even during predition.

That's why there is an entire range of such weakness that doesn't affect AMD while it hurts Intel very strongly.

Re: AMD Ryzen 3000 announced

#346

Post by pdimitar »

jwr wrote: post #285
mfatica wrote: post #281

Earlier quoted context omitted.

Single threaded performance is irrelevant at this point

On the contrary, it is absolutely crucial for some of us (yes, even for developers). I work with an interactively compiled language (Clojure and ClojureScript) and while I rarely do full "builds", I care a lot about recompiling single files and reloading them in the browser. That is not a multi-threaded job and requires single-threaded performance. This is why when looking at iMacs, I'd rather get the iMac than the i…

The iMac Pro's Xeon CPUs have pretty good turbo boost frequencies. If your machine isn't pegged on all cores when you want to do the task specified in your post then you should have zero problems.

Re: AMD Ryzen 3000 announced

#347

Post by saltminer »

sebazzz wrote: post #345
saltminer wrote: post #229

Did AMD talk about VME instructions? I'd love to be able to run a Win 9x VM again.

Running a VM is already possible. Graphics acceleration is what the real issue is.

I'm referring to virtual 8086 mode enhancements. They're broken on Ryzen 1- and 2-series chips, AMD responded to my bug report saying they had no intention of fixing them in microcode updates. You wouldn't notice this unless you tried to run older OSes in a VM, like DOS-based Windows, or some applications in DOSBox.
Post reply on HN