WÒOW..just think where we were 3 months ago with desktops.... and tosay ryzen prices just FELL 23%, that sandwich sized chip (have you seen the size of this thing!?!?) is imminent, and after years of waiting I can finally buy my project pc come June 20th. dual cpus cant wait....actually maybe will wait.. im not an early adopter. Amd! Amd!
http://wccftech.com/amd-ryzen-7-prices- ... er-launch/
http://wccftech.com/amd-ryzen-threadrip ... -platform/
AMD's a-comin!!!... and Intel's been a-dunnin!!
-
fluffy_little_something fluffy_little_something https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=281847
- Banned
- 12880 posts since 5 Jun, 2012
I don't know if such early price drops are a good idea.
There is no point in a price war as they have to earn some money. And I think people are willing to pay a decent price for a great product.
And they might be cannibalizing their own Ryzen 7 lineup when they launch Ryzen 9 so soon after Ryzen 7.
I suppose that is also why they gave Ryzen 7 only two memory channels, which many people found odd at the time. AMD were already finishing off Ryzen 9 with quad-channel memory support for the the very top segment when everyone thought Ryzen 7 was the top.
But I am not really happy about that flood of awesome new processors as they might lead developers of music software to neglect CPU efficiency, forcing people to buy new computers, who can't afford to spend 1000 to 1500 bucks (after all you need a whole new platform, not just the processor). And at the same time forcing people to switch to W10.
There is no point in a price war as they have to earn some money. And I think people are willing to pay a decent price for a great product.
And they might be cannibalizing their own Ryzen 7 lineup when they launch Ryzen 9 so soon after Ryzen 7.
I suppose that is also why they gave Ryzen 7 only two memory channels, which many people found odd at the time. AMD were already finishing off Ryzen 9 with quad-channel memory support for the the very top segment when everyone thought Ryzen 7 was the top.
But I am not really happy about that flood of awesome new processors as they might lead developers of music software to neglect CPU efficiency, forcing people to buy new computers, who can't afford to spend 1000 to 1500 bucks (after all you need a whole new platform, not just the processor). And at the same time forcing people to switch to W10.
-
- KVRAF
- 1929 posts since 4 Nov, 2004 from Manchester
No, the are various things attributed to RAM issues, then the are RAM issues they have fixed and the are RAM issues that are still awaiting fixes.incubus wrote: Now just hold on one cotton-picking minute......in this (and every single other thread about this) there is mention of ram issues. Now there isn't?
I can't keep up.
I'm just asking which one specifically you were referring too!
A lot of it is just controller and IC incompatibilities. These happen with any new chipset (seriously, X79 made me lose my hair) and are normally nailed out over time. Problem is they are just as likely to crop up on any new chipset that follows, so I was a bit confused when I saw suggestion that it'd be ironed out by then when it's a new chipset and new rules.
Or I was wondering if it was in reference to something a bit more uncommon. I saw comment earlier in the thread that memory bottlenecks are often a problem they try and work around with in cache which is perfectly correct. It does also depends on how the cache is arranged and optimized and I still think some of the performance questions are coming from poor cross addressing between cores via cache implementation.
Remember back when Ryzen launched the was the suggestiong by AMD that the Windows thread handling wasn't optimized and then they backpedaled and said that it was "working as intended" and then said little more.
PCPER ran a piece on it here : https://www.pcper.com/reviews/Processor ... ver-Bullet
The closing statement echos something I noticed too.
"In the process, we did find a new source of information in our latency testing tool that clearly shows differentiation between Intel's architecture and AMD's Zen architecture for core to core communications. In this way at least, the CCX design of 8-core Ryzen CPUs appears to more closely emulate a 2-socket system. With that, it is possible for Windows to logically split the CCX modules via the Non-Uniform Memory Access (NUMA), but that would force everything not specifically coded to span NUMA nodes (all games, some media encoders, etc) to use only half of Ryzen. How does this new information affect our expectation of something like Naples that will depend on Infinity Fabric even more directly for AMD's enterprise play?"
This was the problem with Bulldozer before it that we all hoped would be fixed and resides in the hardware being largely considered two different chips for addressing by the code due to the way the CPU is designed and the cache optimized. They carried it over to Ryzen annoyingly and right now I'm waiting to see a chip schematic to see if it also applies to thread ripper.
I pretty much attributed this for the inconsistent CPU performance at low buffer settings I saw and it's also why I don't expect that to be fixed without a chipset overhaul. Threadripper might have it, but then Threadripper is a server platform and the wouldn't be anything to gain by AMD in changing it for that market.
Xeons also have this problem when using multiple chips and they tend to take an overhead hit when more than one is in the system, which is one of the reasons I don't consider them good value as the regular chips don't have the same overhead to deal with ("Soft vs Hard NUMA addressing" if you want to read more about it)
Just as a closing note, the method of address they are using is the correct one for servers and handling vast amounts of none real-time data, and I completely understand why they've implemented it this way. The fly in the ointment is trying to address the cores quickly enough in real-time when dealing with memory calls across the various cores, and doing it quickly enough that none of the audio gets discarded due to speed of the buffer reset... thus explaining (I think) why the chips currently are more load efficient at higher buffer rates.
-
fluffy_little_something fluffy_little_something https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=281847
- Banned
- 12880 posts since 5 Jun, 2012
Multithreading?
-
- KVRAF
- 2945 posts since 23 Dec, 2002
I can't see this as anything but good for musicians. After a 4 year rut we are getting some significant power increases. Bring on the price reductions and core count and let's hope AMD and Intel truly go at on all fronts. You can choose whatever platform you want because both will have been made better by the other. Good times!
fluffy_little_something wrote:I don't know if such early price drops are a good idea.
There is no point in a price war as they have to earn some money. And I think people are willing to pay a decent price for a great product.
And they might be cannibalizing their own Ryzen 7 lineup when they launch Ryzen 9 so soon after Ryzen 7.
I suppose that is also why they gave Ryzen 7 only two memory channels, which many people found odd at the time. AMD were already finishing off Ryzen 9 with quad-channel memory support for the the very top segment when everyone thought Ryzen 7 was the top.
But I am not really happy about that flood of awesome new processors as they might lead developers of music software to neglect CPU efficiency, forcing people to buy new computers, who can't afford to spend 1000 to 1500 bucks (after all you need a whole new platform, not just the processor). And at the same time forcing people to switch to W10.
-
- Banned
- 5357 posts since 7 May, 2015
@ ram.
Again, I bow to Kaine's knowledge, but I can't keep up.
Hope they get everything sorted so that people can simply enjoy the product, AND/OR, you are enjoying the product. (ymmv)
Again, I bow to Kaine's knowledge, but I can't keep up.
Hope they get everything sorted so that people can simply enjoy the product, AND/OR, you are enjoying the product. (ymmv)
-
- KVRAF
- 1603 posts since 18 Feb, 2005 from Serbia
Already enjoying it.incubus wrote:Hope they get everything sorted so that people can simply enjoy the product, AND/OR, you are enjoying the product. (ymmv)
It's easy if you know how
-
- KVRAF
- 1603 posts since 18 Feb, 2005 from Serbia
It's easy if you know how
-
- KVRian
- 700 posts since 19 Jan, 2008
Seems Intel is falling behind, they don't know what to do and rush with stuff that aren't ready to compete with Ryzen.
https://www.youtube.com/watch?v=TWFzWRoVNnE
https://www.youtube.com/watch?v=QJNRtGo5IMc
And there are a lot of videos like these on youtube, everyone is questioning Intel's decisions which are really dumb ones. Introducing "hardware DLCs" like for other RAID than 0 and other stuff like that. Also let's not forget their prices are way too high to compete with AMD's platform.
https://www.youtube.com/watch?v=TWFzWRoVNnE
https://www.youtube.com/watch?v=QJNRtGo5IMc
And there are a lot of videos like these on youtube, everyone is questioning Intel's decisions which are really dumb ones. Introducing "hardware DLCs" like for other RAID than 0 and other stuff like that. Also let's not forget their prices are way too high to compete with AMD's platform.
-
- KVRian
- 826 posts since 25 Aug, 2006
Really interesting the way the Core i7-6800K lags behind the 1800x in the dsp test but thoroughly dominates the virtual instrument "instant high demand" test. Right now the i7 is about $110 cheaper.
-
fluffy_little_something fluffy_little_something https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=281847
- Banned
- 12880 posts since 5 Jun, 2012
The Ryzen 1700 is not much slower than the 1800X, but a lot cheaper. But it has not been included in the ranking...
-
- KVRAF
- 1603 posts since 18 Feb, 2005 from Serbia
It's easy if you know how
-
- KVRAF
- 1603 posts since 18 Feb, 2005 from Serbia
