I always took it to mean the top SKU being at least 3.4 GHz base clock, which seems to be correct.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
*** AMD "Zen" thread (inc AM4/APU discussion) ***
- Thread starter Boomstick777
- Start date
Soldato
Does base clock really matter to some? You can expect overclocking in the region of 4Ghz so why does it matter.
Agreed, with all the chips being unlocked most here won't care about base clocks. It's still interesting as a comparison to what Intel offers though, especially as Intel is getting pretty close to the bone in terms of overclocking headroom.
Caporegime
No she said at least 3.4ghz for Ryzen - not the top bin.
The whole issue is look at the top bin SKU in that list it says 3ghz base clockspeed with a 3.6ghz Turbo.
BASE clockspeed.
It's a rubbish fake which all the rumour sites are jumping on and then they will have another article later saying how AMD magical has raised clockspeeds.
More clickbait rubbish.
Edit!!
Watch the presentation again she said Ryzen would be shipping with AT LEAST a 3.4ghz clockspeed.
Jeez,why are alternative facts so popular now?
That chart does not say 3.4GHZ+ base clockspeed even for the top SKU.
It says 3.0ghz base clockspeed which means Lisa Su lied according to you all.
Last edited:
Soldato
I would say that a higher base clock, considering the TDP is fixed would look better for overclocking further. If AMD can pull an additional 100-200Mhz out of it without increasing the TDP then I would say it looks better for the chips ability to overclock.
Last edited:
That's still open to interpretation. Again, given the context of talking about an unfinished product, I took that to mean that they've achieved 3.4 GHz and looking to go higher. In other words, they know they can hit 3.4 GHz at least, which in turn means their highest SKU will be at least 3.4 GHz base clock.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Jesus, when talking about an architecture, you talk about what it can achieve. No one ever when talking about the architecture and not specific SKUs would ever talk about anything but the highest clocked SKU, the flagship.
What the chip is capable of is what you're advertising, not what every single model will be.
You realise there will be lower power versions and maybe laptop versions, you think they'll all be clocked the same. You think Intel would give an architecture talk on Skylake and say all chips will be higher clocked than.... and give the lowest clocked for the lowest power device to ship in a laptop?
Su was talking about the flagship, because this represents what the architecture is capable of achieving, if you think there won't be lower clocked chips than the flagship, you're off your rocker.
If you see a slide which says 8core/16 thread and 3.4Ghz, when talking about and architecture and with no list of various chips, it means there will be AT LEAST ONE chip, and definitely the flagship chip, that will meet these specifications. Nothing more or less, at no time in the history of chips does stating the expected clock speed mean every chip will surpass those expectations.
Think about it for a second, the slide says 8core/16thread, does that mean... every Zen comes with at least 8 cores and 16 threads, does that make sense? Or is that the maximum we'll see, which makes more sense in that context?
More to the point, lets say you have 3.6Ghz base clock max model Skylake, and 1.2Ghz base clock 10W models.... do you run around saying SKylake will be 1.2Ghz +, or do you say 3.5Ghz+(presuming at that stage 3.6Ghz is max you've achieved and may not be final). It would be utterly absurd to use the 1.2Ghz figure and completely normal to give the highest clock of the flagship model.
What the chip is capable of is what you're advertising, not what every single model will be.
You realise there will be lower power versions and maybe laptop versions, you think they'll all be clocked the same. You think Intel would give an architecture talk on Skylake and say all chips will be higher clocked than.... and give the lowest clocked for the lowest power device to ship in a laptop?
Su was talking about the flagship, because this represents what the architecture is capable of achieving, if you think there won't be lower clocked chips than the flagship, you're off your rocker.
If you see a slide which says 8core/16 thread and 3.4Ghz, when talking about and architecture and with no list of various chips, it means there will be AT LEAST ONE chip, and definitely the flagship chip, that will meet these specifications. Nothing more or less, at no time in the history of chips does stating the expected clock speed mean every chip will surpass those expectations.
Think about it for a second, the slide says 8core/16thread, does that mean... every Zen comes with at least 8 cores and 16 threads, does that make sense? Or is that the maximum we'll see, which makes more sense in that context?
More to the point, lets say you have 3.6Ghz base clock max model Skylake, and 1.2Ghz base clock 10W models.... do you run around saying SKylake will be 1.2Ghz +, or do you say 3.5Ghz+(presuming at that stage 3.6Ghz is max you've achieved and may not be final). It would be utterly absurd to use the 1.2Ghz figure and completely normal to give the highest clock of the flagship model.
Last edited:
Soldato
Think you need to watch the presentation to understand where their confusion originates from....
Associate
If you look at the clockspeed column it's not suggesting it's showing base to max turbo or anything, it's suggesting possible base speed ranges. All of those ranges meet or exceed 3.4GHz so strictly it could all be accurate without contradicting Lisa Su.
No it doesn't, see above. Base frequency range means the range in which the base frequency will sit. Likely varying depending on model e.g. a 1700 isn't likely to have the same base speed as the 1800X
Last edited:
Soldato
6c/12t listed to compete with the latest i5 
Soldato
6c/12t listed to compete with the latest i5
Maybe they mean in terms of price rather than performance? Or maybe it's all made up.
Last edited:
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Again, the way he's interpreting it, if that slide means that EVERY Zen has the same specs, then EVERY Zen released is a 8 core 16 thread... that interpretation flies in the face of any logic, every fact known about what is coming and makes no sense.
There is also 20+ years of cpu architecture launches that says when talking about a chip the flagship is the 'peak' of an architecture and what is spoken about in these cases every single time.
When Nvidia first spoke about Fermi, it was about the biggest Fermi, when Nvidia first spoke about Pascal, it was about the biggest Pascal, when AMD first spoke about Vega, it was about biggest Vega(specs, bandwidth, etc).
This is a 100% industry standard. Go on Qualcomm's site (as I have recently) and it lists only the max clock speed achievable for that chip ... yet dozens of phones come with lower clocks. When Micron talked about gddr5x, 14Gbps was talked about only... yet it launched with 10Gbps. When HBM2 was talked about, 2Gbps was mentioned, yet the first chips available are 1.6Gbps.
In talking about new types of memory, gpus, cpus, ssds or literally anything in the industry, the 'biggest', or top version, or fastest possible designed for version is what is talked about, because that is what the technology is capable of, there will always be smaller, lower binned, salvaged or derivative versions available.
Soldato
As it says, it will "offer an 8 core chip with at least 3.4GHz base clock" not that all chips will be 3.4GHz for the 8 core chip. You have jumped to a conclusion that it meant all.
It also stated that AMD isn't releasing it's absolute minimum. It appears you are cherry picking what is being said and shown.
Also the chart is the base clock which the top SKU is actually 3.6GHz. There are a lot of people who cannot read here and it is frustrating.
Soldato
Right the slide shows that the base clock as follows to clarify:
R3 series will have base clock ranging from 3.1-3.4GHz
R5 series will have base clock ranging from 3.2-3.5GHz
R7 series will have base clock ranging from 3.0-3.6GHz
So with that it is then that each one will offer different steps within the range. The R5 series is split in two as they are offered at 2 different core arrangements but with same clocks depending on model.
The Pro series looks to be supporting ECC (which contradicts what I was told a month ago but has more weight to it) and will be the slower speed with the the standard & X versions be non-ECC supporting but offer higher base clocks with of course the X being the enthusiast model.
So with that and what she said I believe she is correct they they are offering Ryzen with a minimum base clock of 3.4GHz and they are even doing that with each range of chips right from the R3 to the R7.
Further to that, the table does not note turbo speed and that is because it is still to be confirmed as they are still working on the Architecture to get the best out of it for this purpose.
People are just not reading the table right and then miss-understanding the context of what she stated. Yes it could do with clarity but I don't see any issue in what has been said and shown as they do tally up.
Anyone else deciding that it is not the case are offering an alternative view that does not tally with what was said or shown.
R3 series will have base clock ranging from 3.1-3.4GHz
R5 series will have base clock ranging from 3.2-3.5GHz
R7 series will have base clock ranging from 3.0-3.6GHz
So with that it is then that each one will offer different steps within the range. The R5 series is split in two as they are offered at 2 different core arrangements but with same clocks depending on model.
The Pro series looks to be supporting ECC (which contradicts what I was told a month ago but has more weight to it) and will be the slower speed with the the standard & X versions be non-ECC supporting but offer higher base clocks with of course the X being the enthusiast model.
So with that and what she said I believe she is correct they they are offering Ryzen with a minimum base clock of 3.4GHz and they are even doing that with each range of chips right from the R3 to the R7.
Further to that, the table does not note turbo speed and that is because it is still to be confirmed as they are still working on the Architecture to get the best out of it for this purpose.
People are just not reading the table right and then miss-understanding the context of what she stated. Yes it could do with clarity but I don't see any issue in what has been said and shown as they do tally up.
Anyone else deciding that it is not the case are offering an alternative view that does not tally with what was said or shown.
Soldato
Not everyone wants to overclock so base does make some sense. What is more interesting is that if all are unlocked, how far can the cheaper chips really be pushed to get closer to their more expensive counterparts.
That is something I am still struggling to see. I understand the locked chips Intel provide because then it gives way to having unlocked at a premium. However if all chips can clock and you can at least get close to the higher costing chips and clock say from 3.0GHz to 3.6GHz for the R7 then if the price is £100 cheaper I can see a lot picking them up.
Are they banking that those people who pay more will then always clock higher via OCing because the chip is a better unit? Can they really bin that many chips to do this? With Intel they take the chips and lock them or don't and thus you don't know if the chip that is locked would have been a good overclocker chip if in the unlocked batch.
Anyone that bought a Kaby Lake i5/i7 knowing how little improvement there was and knowing that Ryzen was 1-2 months away was probably a bit dim anyway.
Depends really, if they were running a Phenom II or Core2 system and didn't want to wait 2 months then it was a fine move.
Because hardly anyone actually overclocks their CPU, the vast majority of K series processors are purchased for their higher clocks rather than their unlocked status.
Soldato
If the workload demands it, the cores have dynamic scaling. So base clock is still not of great importance.
Whilst on that topic, I've heard this works in a fairly similar manner to ASUS' thermal control tool implemented on KBL and BWE...
Soldato
Yep I understand that but then the dynamic scaling always went from the base clock thus if it was a chip at 3.0GHz then it could dynamic scale (turbo) to say 3.4GHz and then the chip that was set at 3.2GHz could dynamic scale (turbo) to 3.6GHz and so forth.
That is basically showing the base clock is the limiting factor (as it is with Intel) depending on how it is set. It makes even less sense if all chips have a base speed but all can turbo as high as one another. It would also mean an even bigger lottery with people buying 10 cheaper chips, trying them all and then selling off/returning the chips they don't want because they have found a golden nugget and that would just inflate prices.
In regards to the thermal control and the system being similar to this, all that is going to do is that a chip that is too hot will not turbo or dynamic scale to it's maximum but it wont suddenly scale up to 4.2GHz as long as it is in the temp range.
This is only really a problem if not using a decent cooler and to protect the chip from a manual clock if you set the voltage etc higher from my view or maybe for builders like Dell & similar will have more use of this.
The scaling isn't fixed to 4 multipliers, it's a quite common choice but the are others too (I.E my laptops i7 has a 2.4 base and 3.5 turbo, my desktops i7 has a 3.4 base and 3.9 turbo).