AMD Threadripper 1950X Trounces Core I9-7900X In Multithreading Benchmark (pcper.com)
dryriver writes: The Cinebench R15 benchmark is a popular tool for measuring how well CPUs cope with multithreaded compute loads. AMD's Threadripper 1950X 16 core CPU, priced at $999 according to AMD, benchmarks 41% faster in Cinebench R15 than Intel's also $999 10 core Core i9-7900X CPU. While Intel's Core i9-7900X scores 2186 points on Cinebench, AMD's Threadripper 1950X scores 3046 points. Even the cheaper 12 core $799 Threadripper 1920X is over 200 points faster in Cinebench R15 than Intel's Core i9-7900X. Intel has its own 16 core Core i9-7960X in the works, performance yet unknown, priced at $1,699, but AMD's 16 core part currently appears to be a full $700 cheaper than Intel's MSRP. It remaines to be seen who is faster in single-threaded performance -- Intel may take that crown --and what the power consumption of a fully loaded Threadripper looks like compared to its Core i9 counterpart.
but it looks like my Xmas present to myself will be a new AMD box.
Pain is merely failure leaving the body
I surely hope it's servers. These processors would be silly in a desktop computer. We're not even fully loading down 2-8 core machines now. Gaming performance has and still is a single core endeavor, and even now, most of my stuff has trouble pegging any cores to 100% for any length of time.
About the only thing I do that consumes a lot of cpu time is compiling. Not very many computer users compile stuff.
Again, it's ultimate more of the same lackluster improvements. Throwing more threads/core at stuff, when it's still who's got the FASTER single core that matters at the end of the day. At least in my opinion.
For servers however, running virtualization stuff, these CPU's should be great, squeeze even more out every physical server unit.
I managed to peg my 8 core Xeon at nearly 100% CPU usage for about 6 months straight - 7 days a week, 24 hours a day doing video transcoding on a library. But yes, any computer with more than 2 cores is really a niche computer these days since 90+ of the people run computers with CPUs idling 90%+. The $100 Ryzen 3 will be more than enough power for the masses. The greatest "performance" boost for personal computers in the last few years -- for the masses -- has been flash based SSDs...
... or will they continue to take us for as much as they can?
the overly affluent getting even more affluence via monopolistic marketing
Two large rounds of layoffs to bring in cheaper h1b visa labor, along with misteps in leadership and innovation made it incredibly easy for AMD to step in. And everyone thought AMD was switching to ARM. Kudos to AMD
The top AMD chip has 40MB of cache which is enough to run an entire Linux distro from cache alone. However, if you use it's virtualization technology then you could have an entire Beowulf cluster on a chip.
It turns out that AMD has been reading my weekly email demands this whole time! ;)
Anons need not reply. Questions end with a question mark.
There's no such thing as real life single threaded these days. The OS is always running a lot of threads and every service uses them. So whether Intel wins a thing that doesn't exist in real life is kinda irrelevent.
I also read a report that Intel is making a single core chip for embedded devices to compete on price. They need a wakeup call.
The processor market is simple, it's 94% ARM based, and 6% AMD/Intel based and Intel's mobile Core i3s are slower than ARM's 8xA53 and A72's.
They need to up their game and lower their price, maximizing profits in a diminishing market niche is suicide.
The ratio of the performance results is 1.3. The ratio of the cores is 1.6. The Intel device is doing more per thread.
2011 called. It wants its joke back.
Wake up! They aren't selling Bulldozer anymore!
...a 40,000 BTU air conditioner for all the heat AMD product creates.
Except if you look at the specs this generation, it's Intel that is using more power and generating more heat.
Kickass multiuser 4k Plex servers on the horizon.
Isnt that computer using windows? Isnt there a potential bug: https://randomascii.wordpress.com/2017/07/09/24-core-cpu-and-i-cant-move-my-mouse/
If its effecting everything, then either CPU might be scoring artificially low points.
Bring on the fix and more benchmarks!
X299 would be just fine if they didn't use a shitty cpu paste. 20 degree drop once you de-lid your $2000 cpu.
does paste really cost that much for them to skimp on it with these? just dont understand the motivation on why they'd do that. last longer?
Going to be comical to see the Intel fanbois spending the next year or so justifying why they bought the slower, more expensive chip. LOL.
The company I work for has a software product that takes roughly 10 minutes to compile. You don't always have to compile everything, but sometimes you do. The developers get the most cores per dollar they possibly can, as every core cuts down on compile time by a couple of minutes, which can save hours over the course of a month.
12 cores for $800? Yes please.
My Other Computer Is A Data General Nova III.
See subject: Increase cpu core count @ hardware level (OS can use it for starters ala this in Windows for example):
Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Executive]
"AdditionalCriticalWorkerThreads"=dword:00000008
"AdditionalDelayedWorkerThreads"=dword:00000008
* I.E. - How much extra cores will help BEYOND today's CPUs for the OPERATING SYSTEM itself (in Critical Worker Threads) in juggling threads in itself & for other processes (in Delayed Worker Threads) per https://msdn.microsoft.com/en-us/library/cc615012(v=bts.10).aspx/
Here I use 8 for an Intel Core I7 as shown above (both in 1st a 920 & currently a 4790k, since they're quad core (& hyperthreaded) & it was lesser based on physical core count of earlier systems I had (this setting has been around since, iirc, Win2k (correct me IF I am off/wrong - it's been SO long since then)...
(Those are settings in WINDOWS you can adjust to take advantage of added cores as you upgrade to CPUs w/ more cores, for example).
ANYTHING/EVERYTHING, in theory, gains there alone (less "process scheduler thrashing" in other words) - I don't care so much about applications/programs (they are probably written to their practical limits anyhow as to what threadwork will gain them) but again, MORE about how the OS will utilize them (per the 2 TUNABLE PARAMETERS in the .reg file I note above as a way to REALLY use the extra cores, almost guaranteed - Windows allows it, not sure of other OS like *NIX based ones).
APK
P.S.=> The rest will be done @ compiler level (already good, only depends on HOW you can leverage it OR if internal-to-program itself datasets AND PROCESSES (imo, a Gannt chart illustrates this well) allow for it - not all do) & it's always that way, pretty much - hardware 1st, software catches up (& it does, mostly inefficiently @ 1st, sucking up the CPU cycles/efficiencies gained)... apk
The article is silent about vectorization, and Intel invests a lot in that lately. Do we know anything about the compilation flags of that copy of cinebench? If not, the assessment could be extremely unfair. A newer set of vectorization instructions corresponds to a longer vector size for arithmetic operations that can be carried out concurrently. For example, in HPC applications, enabling the highest available level of AVX can lead to 2x gains compared to code compiled for legacy systems.
Seriously? We're reading news that a 16 or 12 core CPU beats a 10 core CPU in benchmarks. This is a sad day for AMD fans.
Intel has serious security problems with their FSP and management engine blobs, which for example break verified boot on ChromeOS such that NSA could implant a persistent rootkit that bypasses ChromeOS's key verification, without holding Google's private key and without physical access to the device.
The way this works is that management engine blobs must exist, or else the CPU reboots a few seconds after startup. While you can check the signature on the blob, you can't check the signature on the blob before it gets control, which is what verified boot requires. It's not just a matter of source code. Although more source code does need to be provided for security, realistically there are some black boxes in the CPU ("microcode", etc.), but this is a writeable black box that can be overwritten with a rootkit on the victim's computer, yet not distributed to all other computers so it can't be studied, and so there isn't a record of their collaboration or incompetence later to hold them accountable. Verified boot resists attacks by at least forcing all computers to be the same: all backdoored or none backdoored, while we know NSA prefers to target individuals (1) because people aware of a leak plug it. unaware targets leak more. (2) for NOBU. known backdoors can be reused by others. (3) because otherwise their relationship with collaborating US companies would be come impossible. IMHO Intel is basically offering up a silver-platter backdoor to them, but the "management engine" and "RAM initialization and stuff" excuses are just plausible enough for credulous neckties to claim deniability and buy Intel.
AMD has all the same problems with their "PSP". Snowden has suggested this is the time for AMD to offer an alternative to this backdoored-by-design US CPU that's built from the ground up with a welcoming, unremovable module perfectly fitting the way US spy agency likes to implant things.
AMD, is there any progress on this? You seem to have backslided since Athlon FX, yet now it's more important than ever to get this situation cleaned up.
I don't think I will ever spend more than $300 on a CPU. It was always enough to buy the fastest for home. What happened?
A percentage of chips are toast in soldering, so it is more expensive.
On a related note, Tom's Hardware called the 7900X "a factory overclocked chip". It generates so much heat at that it needs water cooling to run without throttling.
http://www.tomshardware.com/re...
16 core CPU beats 10 core CPU. Well thank you Captain Stupid.
This story is a total dupe. Slashdot editors aren't paying attention these days... https://hardware.slashdot.org/... The Cinebench scores are referenced and linked right in there.
For now. As compute resources become available mainly through extra cores, programmers will invest more time in threading and parallel processing.
Even now, programmer can depend upon single stream logic and get hardware speedups. Those gains are clearly getting smaller each year but they still appear. Single thread logic is simply easier to write and all programmers have the relevant knowledge and training.
What I'm saying is, programmers who simply must get faster performance out of their code are facing pressure to move to efficient parallel design practices. Eventually many or most programmers will get the message and it will become Standard Operating Procedure. It isn't yet but it will come.
Nee features coming t games are for animations that likely use the cpu. They all the animation to slip/grab things. I.e. one animation that all characters use but is adjusted on th fly to attach a hand to a gun exactly where it needs to. Or feet/characters to get ouch the terrain and change positioning n relatively to it n a human way. There are other e,xamples but this would likely run on the cpu.