New Ryzen 2 (Pinnacle Ridge) gets only 200 MHz boost according to a leak

With DDR4 ram price where it is, I will wait until Zen2, especially since it is sampling now with Meltdown and Spectre silicon fix. Still sporting an OC'd FX8370e in my gaming rig. It is holding up on multitreaded games surprisingly!
 
The FX series is fine for gaming at resolutions above 1080p.. at 1440 and 4k, there isnt much difference in frame rates.. compared to even the newest model intels..
 
KedsDead said:
The FX series is fine for gaming at resolutions above 1080p.. at 1440 and 4k, there isnt much difference in frame rates.. compared to even the newest model intels..
Click to expand...

This is exceedingly game dependent, to the point of basically being false. Any game that depends heavily on single-thread performance, or just CPU performance period, will very likely dog on an FX CPU.
 
IdiotInCharge said:
This is exceedingly game dependent, to the point of basically being false. Any game that depends heavily on single-thread performance, or just CPU performance period, will very likely dog on an FX CPU.
Click to expand...

That is a true statement for games that run at 1080p.. where they are more CPU dependdnt.. at 1440p and higher, most of the games out there are more GPU dependent.. the difference is more noticeable at 4k where the load is shifted to the GPU..
 
KedsDead said:
That is a true statement for games that run at 1080p.. where they are more CPU dependdnt.. at 1440p and higher, most of the games out there are more GPU dependent.. the difference is more noticeable at 4k where the load is shifted to the GPU..
Click to expand...

This is also exceedingly game dependent. It's more true, but only due to the balance of performance demands between the most demanding games and the resolutions most used for gaming. It's a curve, and it will shift over time.

And it's also a general starting point for performance discussion, not a conclusion. Some games (and/or game types) are more dependent on the CPU for frame-times even as average framerates may climb. This is why CPUs with faster single-thread performance are desirable for gaming.
 
IdiotInCharge said:
Your argument is invalid. You're using cost to determine pricing, when the two are only very loosely related- something you seem to understand yet choose to ignore in your non-rebuttal.

Maybe a little education will help. You know, before every computer processor price range from every significant manufacturer in the history of CPUs is listed as an example.
Click to expand...

Damn, you're opposing him using his own arguments. What's up with that?
 
dgz said:
Damn, you're opposing him using his own arguments. What's up with that?
Click to expand...

The original point was "Pricing is a nonlinear function of performance. E.g. getting 15% more IPC from A CPU requires about 30% more transistors in the design. So the faster chip will cost at least 30% more..." So the IPC/cost ratio is worse. Then smilr replied with an invalid argument that ignored what I said.
 
juanrga said:
A list of differences between Skylake and Kabylake was given in #139 including the new graphics microarchitecture. Everyone agrees that Kabylake and Skylake have the same CPU cores, despite Kabylake uses a different process node (14nm+) and has improvements as Speed Shift v2. However, AMD guys pretend that Pinnacle Ridge has different cores than Summit Ridge only by using 14nm+ and having Precision Boost 2.

Not only I am using the same standard for both Intel and AMD, but those 'innovations' that are attributed to Zen+ are already present in the Zen cores in Raven Ridge.

The chip leaked is a qualification sample. Usually qualification samples have the same specs than the commercial processors. E.g. the qualification samples for the 1800X/1700X/1700 had the same clocks than the final processors you can get in stores, as demonstrated in this thread.

It is true that the CPU clocks for Raven Ridge qualification samples differ from the clocks for the final processors. However, those processors are APUs. A minimal change/improvement in the iGPU side, can reduce the power consumption and leave more thermal room for increasing the clocks on the CPU side. This freedom is not available on Pinnacle Ridge.

The leaked qualification sample for 6C/12T Pinnacle Ridge brings 200MHz extra over the corresponding 6C/12T Summit Ridge processor.

The top 4C/8T Summit Ridge has 3.5/3.7GHz clocks. The top 4C/8T Raven Ridge has 3.6/3.9GHz clocks. Is it causality that the 14nm+ chip also runs at 100/200MHz higher than the 14nm chip?

All info available points to Pinnacle Ridge processors having about 6% extra MHz for base/boost and about 12% extra MHz for XFR compared to Summit Ridge analogs. Higher clocks for Pinnacle Ridge processors cannot be totally ruled out, of course, but 200--400MHz extra seems a good estimation.
Click to expand...

That's a lot of talking for missing the point. Point: I don't care if they call it Zen, Zen+, or Rosie O'Donuts left ass cheek. The name is meaningless.

As for the performance, I want to see the clocks for a final X-branded CPU before I buy that 200MHz boost is all we'll see. Like I said, they might be sandbagging the 2600 a little, so as not to exceed the 1600X, and to create more of a spread between X-branded and non X-branded product lines (as some others have speculated). Then again, maybe not. But the point is, the specs for the 2600 QS don't really tell us either way.
 
2400g vs 1400 is 400mhz jump at same price point, we'll see if it holds up for other 2xxx ryzens.
 
juanrga said:
And why would we compare the clocks of the top 4C/8T APU model to the 1400 instead to the top 4C/8T CPU, the 1500X?
Click to expand...

Because it's not a 2500x? And it's not the same price? They obviously positioned it as the replacement for the 1400 as it's same price and nomenclature.
 
bobzdar said:
Because it's not a 2500x? And it's not the same price? They obviously positioned it as the replacement for the 1400 as it's same price and nomenclature.
Click to expand...

Except that Raven Ridge 2400g is not a replacement for the 1400. The replacement for the Summit Ridge 1400 will be the Pinnacle Ridge 2400 or whatever it is named.
 
Last edited:
juanrga said:
Expect that Raven Ridge 2400g is not a replacement for the 1400. The replacement for the Summit Ridge 1400 will be the Pinnacle Ridge 2400 or whatever it is named.
Click to expand...

And it will likely be identical to the 2400g but without the graphics. And at a lower price than either the 2400g or the 1400. So yeah, given they're the same price and both x400's, I think it's a perfect comparison. You get 400mhz more clock speed and an igp for the same price on zen+. I kind of doubt, given that, that they'll even bother making a 2400 non g. No point.
 
bobzdar said:
And it will likely be identical to the 2400g but without the graphics. And at a lower price than either the 2400g or the 1400.
Click to expand...

Identical? Nope. I expect double the L3 cache per core and a 2+2 core configuration for Pinacle Ridge. Also I don't expect lower price.
 
juanrga said:
Identical? Nope. I expect double the L3 cache per core and a 2+2 core configuration for Pinacle Ridge. Also I don't expect lower price.
Click to expand...

Agreed on the 2+2 configuration. Disagree on the price. I'd expect at least a nominal drop (~$20-$40) compared to the APU version, otherwise why bother with the non-APU version at all? The extra cache doesn't make much of a difference.
 
DuronBurgerMan said:
Agreed on the 2+2 configuration. Disagree on the price. I'd expect at least a nominal drop (~$20-$40) compared to the APU version, otherwise why bother with the non-APU version at all? The extra cache doesn't make much of a difference.
Click to expand...

Why bother with the non-APU version at all? Because you have a bunch of broken dies in the production line anyway.

Extra cache, soldier vs TIM, non-capped PCI lanes,...
 
juanrga said:
Why bother with the non-APU version at all? Because you have a bunch of broken dies in the production line anyway.

Extra cache, soldier vs TIM, non-capped PCI lanes,...
Click to expand...

I know why AMD would bother. I'm saying there's not much reason for the consumer to buy the non-APU over the APU unless there's a price difference. The extra cache is minor. Same with solder vs TIM (at this price point), and non-capped PCI lanes. All that is going to add up to a couple %. And the solder vs TIM isn't even a benefit to the regular user, only the enthusiast. No, I fully expect a price difference. Not huge, because this is already low, but some.
 
This gives good hope that AMD will be reasonable with 2300x and 2500x prices. They will have a hard time selling those for much over $99 and $169, even with higher clocks since they do not have a backup gpu.
 
bobzdar said:
And it will likely be identical to the 2400g but without the graphics. And at a lower price than either the 2400g or the 1400. So yeah, given they're the same price and both x400's, I think it's a perfect comparison. You get 400mhz more clock speed and an igp for the same price on zen+. I kind of doubt, given that, that they'll even bother making a 2400 non g. No point.
Click to expand...

It is unlikely to be the same as Raven Ridge is based off Zen cores. So it stands that the "2500" will be the likely replacement of the 1500 family. The performance may be more or less in a similar ballpark but Raven ridge is not pinnacle ridge.
 
Nightfire said:
This gives good hope that AMD will be reasonable with 2300x and 2500x prices. They will have a hard time selling those for much over $99 and $169, even with higher clocks since they do not have a backup gpu.
Click to expand...

Adjustments may only take place at the release date. AMD will segment again to ensure maximum profits and so they should as financials show.
 
OrangeKhrush said:
It is unlikely to be the same as Raven Ridge is based off Zen cores. So it stands that the "2500" will be the likely replacement of the 1500 family. The performance may be more or less in a similar ballpark but Raven ridge is not pinnacle ridge.
Click to expand...

Raven Ridge and Pinnacle Ridge use the same cores, because Zen+ is a marketing rebrand of Zen. That cache latency reduction, Precision boost 2, improved memory controller, and other improvements people associates to Zen+ cores are already present on Raven Ridge.
 
Juangra, let's not continue this circular argument. You know what he means. Zen+ is expected to have higher clocks. AMD just confused everything by calling them a 2200g and 2400g when they should have been a 1350g and 1550g or something.

OrAnGe, I don't see AMD adjusting the RR prices when zen+ is released. More than likely, they will launch a 2300x and 2500x that can clock higher with no iGPU but are still the same price.
 
Nightfire said:
OrAnGe, I don't see AMD adjusting the RR prices when zen+ is released. More than likely, they will launch a 2300x and 2500x that can clock higher with no iGPU but are still the same price.
Click to expand...
Probably $10 more, because that's how they do.
 
cant wait for some proper info on this CPU zen+.. i would expect more from AMD tho. it would be in their best interest if possible to do the most they can. cuz intel isnt miles away atm, and it would be bad tactics to pull an intel at this point.
 
Last edited:
Nightfire said:
Juangra, let's not continue this circular argument. You know what he means. Zen+ is expected to have higher clocks. AMD just confused everything by calling them a 2200g and 2400g when they should have been a 1350g and 1550g or something.
Click to expand...

Raven Ridge is 2000 series because it uses the same cores and same process node than Pinnacle Ridge 2000 series (despite marketing rebrands). The difference is on the G meaning graphics. Raven Ridge already has higher clocks; 100--300MHz higher than quad-core Summit Ridge and it overclocks up to 4.2GHz thanks to the new process node.

Is a causality that Raven Ridge gets 200MHz extra whereas the leaked qualification sample for Pinnacle Ridge "gets only 200MHz extra" as stated in the title of this thread?

EDIT: Just looking for other stuff I found this page from AMD, where the marketing department forgot to rewrite Zen as Zen+ :)

  • The "Zen" core, currently shipping in Ryzen desktop and mobile processors, is in production at both 14nm and 12nm, with 12nm samples now shipping.
Click to expand...
 
Last edited:
juanrga said:
Raven Ridge is 2000 series because it uses the same cores and same process node than Pinnacle Ridge 2000 series (despite marketing rebrands). The difference is on the G meaning graphics. Raven Ridge already has higher clocks; 100--300MHz higher than quad-core Summit Ridge and it overclocks up to 4.2GHz thanks to the new process node.

Is a causality that Raven Ridge gets 200MHz extra whereas the leaked qualification sample for Pinnacle Ridge "gets only 200MHz extra" as stated in the title of this thread?

EDIT: Just looking for other stuff I found this page from AMD, where the marketing department forgot to rewrite Zen as Zen+ :)
Click to expand...

The "quotes" around "Zen" refer to a grouping of products that share the name of Zen. Since Zen+ does share some of it's name and architectural ancestry from Zen it would fall under the "Zen" family. They specify in the full quote that this grouping includes the 14nm products and the upcoming 12nm products.
 
juanrga said:
Raven Ridge is 2000 series because it uses the same cores and same process node than Pinnacle Ridge 2000 series (despite marketing rebrands). The difference is on the G meaning graphics. Raven Ridge already has higher clocks; 100--300MHz higher than quad-core Summit Ridge and it overclocks up to 4.2GHz thanks to the new process node.

Is a causality that Raven Ridge gets 200MHz extra whereas the leaked qualification sample for Pinnacle Ridge "gets only 200MHz extra" as stated in the title of this thread?

EDIT: Just looking for other stuff I found this page from AMD, where the marketing department forgot to rewrite Zen as Zen+ :)
Click to expand...

The only thing Raven Ridge had different is precision boost 2.0. It is NOT the new node. The 1900x was also able to hit these higher clocks, so that proves nothing. RR mist be from the same quality of silicon.

We know what you are trying to do. You are trying to convince people that this is what kind of clocks are expected of Zen+. Give it a rest.
 
juanrga said:
Raven Ridge and Pinnacle Ridge use the same cores, because Zen+ is a marketing rebrand of Zen. That cache latency reduction, Precision boost 2, improved memory controller, and other improvements people associates to Zen+ cores are already present on Raven Ridge.
Click to expand...

If it has differences then it can't be the same. Raven Ridge is based of summit ridge with a single CCX, while pinnacle ridge will for all intents and purposes be what kaby was to skylake. Gains made resemble more a step due to x number of factors.

Given its performance is a tick up on the 1500X and comes cheaper locally for me it is quite an intriguing buy.
 
Dermac said:
The "quotes" around "Zen" refer to a grouping of products that share the name of Zen. Since Zen+ does share some of it's name and architectural ancestry from Zen it would fall under the "Zen" family. They specify in the full quote that this grouping includes the 14nm products and the upcoming 12nm products.
Click to expand...

The quotes are also used for "Vega" and "Zen2", but there is no Zen2+ neither Vega+. They use Zen to denote cores on 14nm and on 12nm because it is the same core on both cases, as demonstrated before in this thread.

Nightfire said:
The only thing Raven Ridge had different is precision boost 2.0. It is NOT the new node. The 1900x was also able to hit these higher clocks, so that proves nothing. RR mist be from the same quality of silicon.
Click to expand...

No. Raven Ridge has precision boost 2, reduced cache and memory latencies with IMC upgraded to 2933MHz, and a new process node that gives higher clocks. The improvements on Raven Ridge were already mentioned in #96. and again in #115, and again in #133...

The 1900X uses the same 14nm process node than the 1800X and rest of the series. The 1900X is Threadripper and got slightly higher clocks via binning. The top 10% of the best dies are reserved for Threadripper, the rest of the production goes for normal RyZen series as the 1800X.

OrangeKhrush said:
When Intel does it it is called an upgrade, when AMD does it, it is called a rebrand.
Click to expand...

It is just the contrary, when Intel does it everyone agrees the cores are the same, when AMD does some people pretends that the core has changed and deserves a new marketing label...But all of this discussion happened before:

juanrga said:
Differences between Kabylake and Skylake processors are summarized in many places: use of a 14nm+ node with different pitch, Speed Shift v2, improved graphics cores, Optane, OPI 3.0,... and about 15% higher clocks, but no one claims that Kabylake has different CPU cores than Skylake. At contrary, everyone agrees that Kabylake and Skylake have the same CPU cores:

Kaby Lake features the same CPU core and performance per MHz as Skykake
Click to expand...

You can visit the Intel section and you can check that no one rebrands Intel 14nm++ as "11nm" and no one rebrands the cores in CoffeeLake as "Kabylake+" cores.

It is AMD which is rebranding the Zen cores in Raven Ridge and Pinnacle Ridge as "Zen+ cores". And it is Globalfoundries which is rebranding 14nm+ as 12nm.
Click to expand...
 
demondrops said:
juanrga. why what how strong will zen+ be?
Click to expand...

Dangerous question to ask, you may get Carizo + 55%.

Anyways Ryzen as a product continues to evolve as the platform matures and developers have access. Take Tomb Raider as the best example, in march reviews were done and Ryzen struggled with Tomb Raider by a substantial amount.

MARCH 2017

index.php


FEB 2018

index.php



That is substantial gain given that no clock enhancements were made or hardware level updates, that is the developers adapting the game engine to better utilize Ryzen and the gains were substantial. Zen+ will feature the kind of advances you'd expect from more of a stepping upgrade, higher clocks, better efficiency, cache clocks higher, faster IMC and all this has x influence on a per application usage basis.
 
juanrga said:
The quotes are also used for "Vega" and "Zen2", but there is no Zen2+ neither Vega+. They use Zen to denote cores on 14nm and on 12nm because it is the same core on both cases, as demonstrated before in this thread.



No. Raven Ridge has precision boost 2, reduced cache and memory latencies with IMC upgraded to 2933MHz, and a new process node that gives higher clocks. The improvements on Raven Ridge were already mentioned in #96. and again in #115, and again in #133...

The 1900X uses the same 14nm process node than the 1800X and rest of the series. The 1900X is Threadripper and got slightly higher clocks via binning. The top 10% of the best dies are reserved for Threadripper, the rest of the production goes for normal RyZen series as the 1800X.



It is just the contrary, when Intel does it everyone agrees the cores are the same, when AMD does some people pretends that the core has changed and deserves a new marketing label...But all of this discussion happened before:
Click to expand...

Raven ridge is 14nm,I can find nothing that denotes it as 12nm (or 14nm++), otherwise I agree it has all of the other zen+ enhancements. Everything I can find says pinnacle ridge will be the first 12nm parts.
 
KedsDead said:
That is a true statement for games that run at 1080p.. where they are more CPU dependdnt.. at 1440p and higher, most of the games out there are more GPU dependent.. the difference is more noticeable at 4k where the load is shifted to the GPU..
Click to expand...


False, games that relies on a powerful CPU will stu stut stutterrr & chugg chugggg on an FX CPU at ANY RESOLUTION.
 
You must log in or register to reply here.
Back
Top