Overclock.net banner

61 - 80 of 99 Posts

·
Registered
Joined
·
846 Posts
Meanwhile, over at Intel their executives are on suicide watch.
 

·
Registered
Joined
·
675 Posts
Naa, they were just buying AMD stock for future retirement plan :).
P.S:
Do we know if this 1.6V goes inside the CPU chiplets or the I/O Chiplet runs on 1.6V and the CPU Chiplets run on lower voltage ?[maybe we should stop thinking in monolithic way?]
 

·
Registered
Joined
·
2,938 Posts
I thought there were demo about RAM running at 5GHz for DDR4 on Ryzen 2? I bet Ryzen 2 can easily go beyond 4000MHz this time.
That's on ryzen using 2 sticks of ram. The max potential speed goes down as you use more chs. On tr 2nd gen i was able to get 8 sticks to boot @ 3200 but it still failed training maybe 1/3 of the time. I don't want to play the chip mc lotto or the 30% training failure. With ryzen 3 supposedly having a ram speed sweet spot of 3600 according to AMD, I really expect to not have training failures with TR3 and using all 4 chs.
 

·
mfw
Joined
·
8,623 Posts
Soon we'll be saying "We need Intel to keep AMD in line"
"AMD has been sandbagging increasing IPC and core count beyond 16/32 on mainstream for ages now!"
 

·
Professional Proletariat
Joined
·
5,671 Posts
"AMD has been sandbagging increasing IPC and core count beyond 16/32 on mainstream for ages now!"
Ikr, what kind of evil corporation artificially segments desktop PCs into "mainstream" and "HEDT"? And what's with all this overpriced Threadripper crap?
 

·
Registered
Joined
·
805 Posts
Ikr, what kind of evil corporation artificially segments desktop PCs into "mainstream" and "HEDT"? And what's with all this overpriced Threadripper crap?


The nerve of moving of to 7nm and putting all their 14/12nm specialized engineers and production lines out of work!


Sent from my iPhone using Tapatalk
 

·
Joined
·
2,327 Posts
The nerve of moving of to 7nm and putting all their 14/12nm specialized engineers and production lines out of work!
Oh, and what's with all those cores? Developers don't have any time to catch up! It's a waste of silicon, I want 6Ghz so I can play Dota at 999fps/720p.
 

·
Professional Proletariat
Joined
·
5,671 Posts
Oh, and what's with all those cores? Developers don't have any time to catch up! It's a waste of silicon, I want 6Ghz so I can play Dota at 999fps/720p.
Those cores are glued together m8. G-L-U-E-D together. Just let that sink in for a second.
 

·
Meep
Joined
·
5,582 Posts
i just said it in another thread, whoever sells us the most lanes @ the lowest cost will get many of our $$$$$

why? cuz of the above. We wanna run 1-2 GPUs @ 16x each + 2-4 NVMe @ 4x each... Why? Cuz some of us like to keep a ton of ultra-fast data on our gaming PC & are tired of the now old & tired SATA SSD raid solutions. ;)

not to mention most of us in the above are @ resolutions where 1-2% CPU nonsense means nothing.
If you just want a ton of lanes, a TR 1900X and ASRock X399 phantom Gaming 6(3x x16; 3x x4; 1x x2) can already be had for less than €500 if you buy at the right time, I don't see 62 lanes getting much cheaper any time soon.
 

·
Registered
Joined
·
3,968 Posts
"AMD has been sandbagging increasing IPC and core count beyond 16/32 on mainstream for ages now!"

Intel have much bigger resources than you might think, if they have to dump their SSD department and others investment and fully focus on coming out with killer CPUs, they will do it.


But first, they need to retrench the bunch of useless and greedy ppl at the top management.
 

·
Registered
Joined
·
846 Posts
My Turkish is non-existent nowadays. Anyone know and want to translate or has it all been covered by sources?

The RAM OC strikes me as something that may affect those scores. However, the clock is not right, and I suspect this is an engineering sample. One of the leaks I saw of this mystery processor long before Dr. Su announced it at E3 had an A0 stepping, which I've concluded meant an ES processor. I've played with Intel ES processors in the past bought in bulk that were A1 stepping and thus not meant for mass production.

Intel have much bigger resources than you might think, if they have to dump their SSD department and others investment and fully focus on coming out with killer CPUs, they will do it.


But first, they need to retrench the bunch of useless and greedy ppl at the top management.
You're quite correct, mate. They have immense resources. They spend several times over AMD's total R&D budget each year for their CPU division let alone all other divisions at Intel. How's that worked out for them the last 8 years? Is the 2600K one of the great processors that stood the test of time, or is it simple a good processor when compared to pitifully churned out product since Sandybridge?

You can use Google's custom search settings to look for Intel 10nm being brought up way back in late 2011, early 2012. I suspect Core and later revivals were just very lucky incidences for Intel and they chose to untap as much performance as they could from an architecture.
 

·
Registered
Joined
·
846 Posts
Highest R15 scores on LN2 below

1950X @ 5370 MHz = 4514cb
https://hwbot.org/submission/3622692_theoverclocker_cinebench___r15_ryzen_threadripper_1950x_4514_cb

2950X @ 5489 MHz = 4758cb
https://hwbot.org/submission/3921708_sampson_cinebench___r15_ryzen_threadripper_2950x_4758_cb

9960X @ 5928 MHz = 5320cb
https://hwbot.org/submission/4168151_bigblock990_cinebench___r15_core_i9_9960x_5320_cb

Compared to:

3950X @ ~5400 MHz = 5344cb


Slightly higher frequency than Zen1 but over 18% higher performance. Slightly lower frequency than Zen+ but over 12% higher performance. Same performance as Intel Core i9-9960X but at 9% lower frequency.
I'm not one to nitpick at others representations, but using percent here feels a bit disingenuous to most consumers even though this is an enthusiast site. As you pointed out, there is a large, significant difference in score from the TR1 chip, however, for the Intel processor, there's over a 400 Mhz difference which shows just how much of an IPC improvement there is with Zen 2 over Intel. I'm not sure if this comes down to using the TSMC process or because 12nm and 14 nm on GloFlo was a highly modified mobile platform on licensed tech from Samsung. Unfortunately, that stuff is way out of the breadth of knowledge I know so I'm open to anyone who knows it to explain why apart from that and the modification of how IF works that AMD managed to best Intel, obviously based on what we know and not third party benches.
 

·
Registered
Joined
·
846 Posts
I keep on hearing that TR3 will eventually come to market, including a 32c with 'monolithic' RAM. AMD however may want to see what the new Intel HEDT line will bring to the table, and probably move TR3 upmarket a bit more... and the 3900X and 3950X are opening important segments and price points already, anyhow.

There's also the fact that between 8c and 16c Ryzen 3k and the lucrative $erver-chip market (7nm Epyc, facing increasing demand), a lot of binned 8c chiplets will already be spoken for. Still, like other posts above, I prefer / need at least quad channel RAM and more PCIe lanes than 3950X offers (ie. partial productivity work w/ extra raid cards, Intel network server cards etc)
The ice maker and water chilling system from any zoo that has a polar bear exhibition?
 

·
Hey I get one of these!
Joined
·
7,508 Posts
The nerve of moving of to 7nm and putting all their 14/12nm specialized engineers and production lines out of work!


Sent from my iPhone using Tapatalk
The funny thing is they didn't put them out of work, the IO dies, x570 chipset, and Zen+ APUs are still all made on 12/14nm. :p

I'm not one to nitpick at others representations, but using percent here feels a bit disingenuous to most consumers even though this is an enthusiast site. As you pointed out, there is a large, significant difference in score from the TR1 chip, however, for the Intel processor, there's over a 400 Mhz difference which shows just how much of an IPC improvement there is with Zen 2 over Intel. I'm not sure if this comes down to using the TSMC process or because 12nm and 14 nm on GloFlo was a highly modified mobile platform on licensed tech from Samsung. Unfortunately, that stuff is way out of the breadth of knowledge I know so I'm open to anyone who knows it to explain why apart from that and the modification of how IF works that AMD managed to best Intel, obviously based on what we know and not third party benches.
He used all percentages correctly, taking both clock speed and score into account for all three comparisons. Nothing about it is disingenuous, it is fact.

AMD covered the changes made in Ryzen 3. Anand goes into depth. Skip to page 6 for arch overview;
https://www.anandtech.com/show/14525/amd-zen-2-microarchitecture-analysis-ryzen-3000-and-epyc-rome


tl;dr, cores are made up of lots of parts. AMD shored up on some of those parts weak points so the core can operate more efficiently and get fed/process data faster, such as expanding cache to reduce cache misses.

It's like fixing that one intersection that causes a traffic jam, just... for data. Yes, now another intersection that didn't used to be a problem is now a bottleneck, but overall traffic is moving faster. They can address those new problems next time and make it even faster. Intel has been doing this for a long time, that is where you see "3% IPC increase", etc. The difference is AMD hit 10-15% (in this case, +18% in Cinebench over Ryzen 1), not the 3-5% Intel has been doing.
 

·
Premium Member
Joined
·
876 Posts
I'm not one to nitpick at others representations, but using percent here feels a bit disingenuous to most consumers even though this is an enthusiast site. As you pointed out, there is a large, significant difference in score from the TR1 chip, however, for the Intel processor, there's over a 400 Mhz difference which shows just how much of an IPC improvement there is with Zen 2 over Intel. I'm not sure if this comes down to using the TSMC process or because 12nm and 14 nm on GloFlo was a highly modified mobile platform on licensed tech from Samsung. Unfortunately, that stuff is way out of the breadth of knowledge I know so I'm open to anyone who knows it to explain why apart from that and the modification of how IF works that AMD managed to best Intel, obviously based on what we know and not third party benches.

Not sure I get your point, I clearly state 9% lower frequency than 9960X but yielding about the same score? 5400/5928 = 0.91 or 91% of the 9960X frequency or 9% lower.
 

·
Overclocking Enthusiast
Joined
·
5,962 Posts
Why not use IF to connect the chipset to the CPU? Instead of using pcie Lanes.
 
61 - 80 of 99 Posts
Top