TechARP published some leaked benchmarks which seem to show the performance of two as of yet unreleased AMD processors the 8-core Ryzen 7 2700X and 6-core Ryzen 5 2600. The benchmarks contrast their performance against the current Ryzen 7 1700X as well as Intel's i7 6700K and the results look good. The new chips outperform their predecessors by a noticeable margin and are able to top the Intel part as well. These leaked benchmarks are all productivity software, so we don't have gaming results nor have we seen these two chips paired with extremely highly clocked DDR4 yet but it does give us a glimpse at performance; assuming these are accurate of course.
"Can't wait to find out how fast the 2nd Generation AMD Ryzen processors are? We present to you – the leaked benchmark results and findings of the AMD Ryzen 7 2700X and Ryzen 5 2600 processors!"
Here is some more Tech News from around the web:
- Fleeing Facebook app users realise what they agreed to in apps years ago – total slurpage @ The Register
- Windows 10 Pro vs. Five Linux Distributions In Various Benchmarks @ Phoronix
- TLS 1.3 is incoming to make web browsing safer and snappier @ The Inquirer
- Turning The Beaglebone On A Chip Into A 3D Printer Controller @ Hack a Day
- Ex-ZX Spectrum reboot man threatens sueball over unpaid invoices @ The Register
- Windows on ARM Benchmarked @ TechSpot
- Reolink Argus 2 Wireless Battery Powered Security Camera Review @ OCC
Why are comparisons being
Why are comparisons being done against I-6xxx series chips which are 2 generations old? Give me a comparison to Coffee Lake so I can decide which to buy based on current generations.
Couldn’t you correlate the
Couldn’t you correlate the approximate performance by comparing the older Intel chip with the newer one? Beside this is a leak.
How about doing the benchmark
How about doing the benchmark with the intel meltdown patch on it so we can do an apple to apple comparison…
Intel finally on Feb 21 2018
Intel finally on Feb 21 2018 released meltdown and spectre patches for its newest cpus.
Very few cpus have had this patch installed at this time….
I believe that any future benchmarks should specify whether the meltdown patch has been installed or not because the patch has significant effects on performance for example responsiveness is reduced by over 10%, gaming and especially storage bandwidth is especially affected 29% ! I am seeing benchmarks of Intel cpus vs new AMD 12 nano Ryzen CPU’s and I bet they are using UNPATCHED Intel machines.
At least they should let us know because of the significant performance hit. Do you agree ?
Search for this read
Read the below concluding remarks an eye opener since most intel CPU,s do not have this patch applied yet.. Let’s compare apples to apples here for once.
https://www.anandtech.com/show/12566/analyzing-meltdown-spectre-perf-impact-on-intel-nuc7i7bnh/4
I am more curious about the
I am more curious about the supposed 8C/16T 9700k coming to the Z370 platform…..
I sure did want to read that
I sure did want to read that TechARP article but not at the cost of my 4 core 8 thread procesor going above 33% utlilzation, and rising, with the browser’s UI slowing to a snail’s pace at the north pole in mid to late winter. WTF is up with that, are they mining something there. It’s obvious that the TechARP article is not sourced from that Canard PC Clouseau-esk review of 2700X on an A320 Motherboard.
It looks, at this too early time, like the 2700X is not too bad but still is not up there fully with Intel’s much more costly SKUs but it’s getting closer for sure. No Use in testing things out on any non 400 series AM4 MB’s though as there is a bit more PCIe 3.0 connectivity on the 400 series chipsets! And it’s not that the overall PCIe lane counts have increased on the 400 series AMD/AM4 chipsets it’s that more of those lanes are now PCIe 3.0 on the 400 series MB’s, rather than the formally PCIe 2.0 lanes on the 300 series Ryzen/AM4 MB’s.
I’m Really not going to trust any non 400 series benchmarking of the 2700X. The firmware is not fully ready yet also no matter what they say. So that’s really going to have to wait until the NDA’s expire.
Inspector Clouseau: Do you have a rheum?
Munich Hotel Clerk: I do not know what a “rheum” is.
Inspector Clouseau: [Checks his translation book] Zimmer.
Munich Hotel Clerk: Ah, a room!
Inspector Clouseau: That is what I have been saying, you idiot. A rheum. [gesturing to the hotel’s dog] Does your dog bite?
Munich Hotel Clerk: No.
[Clouseau bends down to pet the small dog; it immediately growls and bites him.]
Inspector Clouseau: I thought you said your dog did not bite!
Munich Hotel Clerk: That is not my dog.
According to AMD’s leaked
According to AMD’s leaked slide the 2700X is more expensive than the 8700k. On its own merit it looks like a good part, but why would you pay for more for something that’s almost as good?
“more expensive” tells me
“more expensive” tells me absolutely nothing I’m wanting a cost per CPU core sort of Cost/Benefit sort of analysis.
“but why would you pay for more for something that’s almost as good” [Disingenuous Alert]
We all see what you did there by the way, so that’s not going to work!
Some folks do more than game and some folks do Cost/Benefit analysis sorts of Look Sees where things like Epyc/SP3 being a better workststion choice than Threadripper/TR4 ever could be on a Cost/Per-PCIe-Lane and a Cost/Per-Memory-Channel bases makes Epyc/SP3 the better workstation deal even with the Epyc/SP3 MB’s costing around $620+(that Gigabyte MZ31-AR0 Single socket Epyc/SP3 MB with 4/x16 electrical and 1 x16 that’s x8 electrical, for a total of 5 PCIe x16 slots). That GB Epyc/SP3 MB has more Features/$ than any TR/TR4 MBs and twice the memory channels(8) and PCIe lanes(128) than any TR/TR4 motherboard SKUs. You see adults do Cost/Benefit analysis sorts of things whereas children and “Adult” Game-Children do not.
Not very much like the Rather Daft and Disingenuous Folks Like Yourself and most gamers in their mother’s basement!
So… because Intel “wins”
So… because Intel “wins” in:
Single threaded, 1080ti equipped, 1080p, looking at max framerate only benchmarks, it’s totally the better chip and AMD MUST be cheaper to compete.
Lets not consider:
Newer games
1440p
4k
Multi-threading
Non-gaming
Gaming on a regular home system
Gaming while streaming
or… you know, any other extremely common scenario. Only the single threaded, 1080ti equipped, 1080p, looking at max framerate only benchmarks win matters.
It’s not good to make
It’s not good to make comments with ignorance as a flag. Ryzen 3xx series chipsets are fully pcie 3.0 compliant. In fact it is stated that x470 is x370 rebranded, but as not every x370 board would be xfr2 enhanced compliant, they decided to rebrand the same chipset so new boards were made by all vendors supporting it, and still some x370 boards will support xfr2 enhanced. In short, 3xx chipsets and 4xx chipsets are the same with xfr2 enhanced and boost difference in every board, no more. Pcie3.0 is present in every ryzen board in the market
It’s nothing to do with the
It’s nothing to do with the 300 series not being PCIe 3.0 compliant it’s more to do with the 400 series replacing more of the PCIe 2.0 lanes with 3.0 capable lanes than about any sort of compliance with any PCIe standard being sub par, how you read that from what was posted is what confuses me.
That Blanket Rebrand nonsense is also not wise as Zen/Zeppelin and Zen+/Zeppelin on GF’s new 12nm node as well. It’s because of the extra time that AMD was afforded for the 2000 series Ryzen CPUs that gave AMD plenty of time for some tweaking on the 400 series ON SOC Chipset IP!
Zen/Zeppelin is an SOC(Northbridge and southbridge on die) so that’s now some Zen+/Zeppelin dies at 12nm with tweaks in that SOC’s(1) northbridge and southbridge IP on the 2000 series SKUs’ Die.
What is really on any Zen/Ryzen Motherboard is up to the MB’s maker and most of the serious IP is included by AMD on the Zen+/Zeppelin SOC’s die with the motherboard makers not having to worry about the NorthBridge/SouthBridge that AMD has provided on the Zen+ SOC. Tweaks have happened as the Zen+/Zeppelin die’s new tapeout at 12nm gave AMD’s engineering Teams the opportunity to tweak even more than what they would have ben able to do if this had just been only a new stepping without any die shrink.
Motherboard makers have had the time to look at any of their 300 series AM4 MB design issues related to traces/layouts and Tweak there also for their 400 series AM4 SKUs. And that Includes fixes/tweaks to any on motherboard controller IP firmware and hardware no part of the Zen+/Zeppelin SOC improvment that AMD undertook.
(1) [See slide 5 of the ISSCC presentation and that slide shows that Zen/Zeppelin is in fact an SOC. And the Ryzen 2000 series are based on a newer Zen+/Zeppelin 12nm shrink with tweaks to the cache subsystem, and tweaks to the Northbridge/SouthBridge ON Zen+/Zeppelin for the 2000 series SOC IP]
“ISSCC 2018: “Zeppelin”: an SoC for Multi-chip Architectures”
https://www.slideshare.net/AMD/isscc-2018-zeppelin-an-soc-for-multichip-architectures
It’s a strong showing and a
It’s a strong showing and a good value / threads proposition but I think we’ll be waiting until 7nm until AMD are truly competitive against intel desktop parts. And that’s just fine by me, I’m happy with my R7 1700 oc ed.
New CPU test on old board
New CPU test on old board using slow ram with no XFR 2.0. Got to watch out for that stuff when checking upcoming reviews.
“Ingore” or “Ignore” in the
“Ingore” or “Ignore” in the headline?
Jsut ingore teh seplling
Jsut ingore teh seplling
Jsut ingore teh seplling
Jsut ingore teh seplling
Typo “ingore”
Typo “ingore”
Oh also disenguine alert! AMD
Oh also disenguine alert! AMD is the best. Hands down. Specter proved this in his recent meltdown benchmark so sorry to tell you.
I see you ignored me, so it
I see you ignored me, so it looks like i won… you really didnt stand a chance to my souperior intelect