AMD sure got good mileage out of Trinity

[Charles Kozierok]

From what I've been able to gather so far, there are no fewer than 1812 specific named CPU lines using the Trinity core. Many of them basically just rebrands of others.

It's bizarre. I get why they're doing it, but still.. wow.

 

[Blandge]

18 specific named CPU lines using the Trinity core

So like A line, R line... There's 16 more of these?

 

[Centauri]

Yeah, I'm confused too.

 

[Charles Kozierok]

A10 for Desktops Series
A8 for Desktops Series
A6 for Desktops Series
A4 for Desktops Series
A10 for Notebooks Series
A8 for Notebooks Series
A6 for Notebooks Series
A4 for Notebooks Series
Athlon X4
Athlon X2
R-Series embedded
FirePro A300 Series
Opteron 6300
Opteron 4300
Opteron 3300

I think that's all of them. You could say all of the "A"s are one series, but AMD actually made all of those different specific product lines.

Also amusing is that many of these are the same chip, rebranded with tiny modifications. The R-Series is 8 chips with weird names like R-268D -- there are four suffixes and I can't find anywhere that says what they are -- and every single one corresponds to one of the An-Mobile chips, perhaps with the GPU speed slightly changed.

The FirePro chips have the same stats as the A10-5000s, just with "pro drivers" and lots of marketing fluff.

Bonus points for recycling "Athlon" with these new chips.. though there never was an Athlon X4 as it turns out. There was an Athlon II X4, and they used that with Llano, but this time they dropped the "II". Why? Who knows.

It's pretty amusing. Makes Core iMumble seem simple by comparison.

 

[podspi]

I was always sad that the Athlon name was diluted as a budget brand. Never understood why they did this (except to copy Intel who did something similar with the legendary Pentium name).

And yes, I agree model numbers are better, but I remember when Athlons and Pentiums were more than just budget lines...

 

[Charles Kozierok]

Actually, I just realized the Opterons don't belong in there, they are not Trinity chips.

 

[Idontcare]

Outside of the FX lineup, the AMD product landscape is as foreign and confusing to me as the Intel product landscape outside of the i7 lineup.

For example even my own laptop, which I've had now for 15 months, I cannot remember the product name of its CPU. Anytime I need to list the processor for specs reasons in the forums I have to go to the "My Computer" screen and remind myself once again what processor model windows thinks I have in my laptop. i5-2XXXM something or other at some clockspeed above 2GHz.

All I can remember is that is it 32nm SB dual-core with HT. Beyond that I can't remember it and I own it and use it everyday. What hope do I have of remembering what the heck an i3-xxxx means in terms of clockspeed, cores, product, etc?

 

[NTMBK]

All I can remember is that is it 32nm SB dual-core with HT. Beyond that I can't remember it and I own it and use it everyday. What hope do I have of remembering what the heck an i3-xxxx means in terms of clockspeed, cores, product, etc?

It's probably a dual core SB with HT, but no boost. Probably

 

[Maximilian]

Outside of the FX lineup, the AMD product landscape is as foreign and confusing to me as the Intel product landscape outside of the i7 lineup.

For example even my own laptop, which I've had now for 15 months, I cannot remember the product name of its CPU. Anytime I need to list the processor for specs reasons in the forums I have to go to the "My Computer" screen and remind myself once again what processor model windows thinks I have in my laptop. i5-2XXXM something or other at some clockspeed above 2GHz.

All I can remember is that is it 32nm SB dual-core with HT. Beyond that I can't remember it and I own it and use it everyday. What hope do I have of remembering what the heck an i3-xxxx means in terms of clockspeed, cores, product, etc?

AMD need an equivalent of the ark.intel site with all the specs listed nicely and minimal marketing BS.

 

[Insert_Nickname]

AMD need an equivalent of the ark.intel site with all the specs listed nicely and minimal marketing BS.

That would be nice... ^_^

 

[lamedude]

This and this not good enough?

 

[Insert_Nickname]

This and this not good enough?

Bookmarked. Thanks... :biggrin::thumbsup:

 

[SocketF]

This and this not good enough?

Lots of errors there ... certainly better than nothing, but compared to ark.intel it's not much.

 

[NTMBK]

Lots of errors there ... certainly better than nothing, but compared to ark.intel it's not much.

Oh yeah, because Intel never lists things like an SDP of 0W. But yeah, Ark seems to be more reliable.

 

[sm625]

Bookmarked. Thanks... :biggrin::thumbsup:

I realized long ago that that site is hopeless outdated. They still havent added ANY 32nm processors!

 

[Charles Kozierok]

As someone who has literally spent dozens of hours researching CPU specs over the last year, let me assure everyone here that Intel is much, much, much better than AMD when it comes to making full stats of their products available.

Sure, you can get basic "speeds and feeds" on most current chips on the AMD site. But anything more than a few years old? Forget it. And the specs they do have are sparse, and often have errors and contradictions.

And the problem isn't even the lack of a proper equivalent of Intel's "ARK". It's the lack of availability of detailed information: datasheets.

Want to know exact details on how memory controllers work, or how PCIe lanes are configured, or pretty much anything else? Intel has datasheets on their site for nearly every chip they've made since the start, detailing every aspect of them in gory detail. AMD? Virtually nothing, and what they do have is hidden in weird places.

People can't even access design datasheets for most of their recent sockets.

My guess is that some executive at AMD decided that this was "proprietary information" and nobody should be able to get it unless under an NDA or something. What is it they are afraid of? I have a hard time understanding how a company in the "challenger" position benefits from specification secrecy when the company they are chasing is so open.

 

[Idontcare]

Oh yeah, because Intel never lists things like an SDP of 0W. But yeah, Ark seems to be more reliable.

Status:
End of Life

That looks to be more like a templating auto-fill default value type situation. Intel must have updated their spec template but did not spend money on salaries for people to go back and vet the auto-generated template for zero-values for products which were alread EOL.

On the other hand it would be nice if AMD would actually fill in the spec values for the FX-8350 I bought, and that AMD still sells, considering it is an FX chip but I have no idea as to the max spec for operating temperature or voltage. Talk about selling something "as is"

 

[Idontcare]

As someone who has literally spent dozens of hours researching CPU specs over the last year, let me assure everyone here that Intel is much, much, much better than AMD when it comes to making full stats of their products available.

Sure, you can get basic "speeds and feeds" on most current chips on the AMD site. But anything more than a few years old? Forget it. And the specs they do have are sparse, and often have errors and contradictions.

And the problem isn't even the lack of a proper equivalent of Intel's "ARK". It's the lack of availability of detailed information: datasheets.

Want to know exact details on how memory controllers work, or how PCIe lanes are configured, or pretty much anything else? Intel has datasheets on their site for nearly every chip they've made since the start, detailing every aspect of them in gory detail. AMD? Virtually nothing, and what they do have is hidden in weird places.

People can't even access design datasheets for most of their recent sockets.

My guess is that some executive at AMD decided that this was "proprietary information" and nobody should be able to get it unless under an NDA or something. What is it they are afraid of? I have a hard time understanding how a company in the "challenger" position benefits from specification secrecy when the company they are chasing is so open.

Companies don't like liabilities and specs are liabilities. Look at the FX-8350 and MSI mobo fiasco. That is a class-action liability, AMD doesn't want that. The less they publish the more plausible deniability they retain for themselves.

Right now buying AMD cpus is a lot like buying chinese made generic goods, good luck finding any specs on the product that you took home from the store. Google all you like, there's nothing official on it from the manufacturer.

 

[Charles Kozierok]

Companies don't like liabilities and specs are liabilities. Look at the FX-8350 and MSI mobo fiasco. That is a class-action liability, AMD doesn't want that.

If true, that's not a great statement of confidence. But I'm not sure I understand what you mean about a class-action liability. Are you referring to this business where the FX-8350 draws more power than spec'ed?

I'm not sure how lack of availability of proper datasheets helps them. If there's an actual lawsuit, all of that information will be subpeonaed anyway, won't it?

Right now buying AMD cpus is a lot like buying chinese made generic goods, good luck finding any specs on the product that you took home from the store. Google all you like, there's nothing official on it from the manufacturer.

Well, there is some stuff out there. And I was shocked that I was actually able to find a summary sheet of key specs on the Piledriver chips when there wasn't one for the original Bulldozer.

But yeah, compared to Intel.. it's pretty sparse.

Oh yeah, because Intel never lists things like an SDP of 0W. But yeah, Ark seems to be more reliable.

SDP is a new spec and not applicable to Xeon chips.

 

[Atreidin]

I don't see why people are still confusing maximum power draw with TDP. They are not the same thing, for AMD or Intel. AMD is not lying when they state a TDP of 125W for their 8350.

I'm not defending the practice but calling it a lie is overdoing it.

 

[Charles Kozierok]

On the other hand it would be nice if AMD would actually fill in the spec values for the FX-8350 I bought, and that AMD still sells, considering it is an FX chip but I have no idea as to the max spec for operating temperature or voltage. Talk about selling something "as is"

Well, according to this, it doesn't even have L3 cache.

I don't see why people are still confusing maximum power draw with TDP. They are not the same thing, for AMD or Intel. AMD is not lying when they state a TDP of 125W for their 8350.

I'm not defending the practice but calling it a lie is overdoing it.

If thermal solutions intended for 125W don't prevent the chip from throttling, isn't that a problem?

 

[Piroko]

The german Wikipedia site has a nice and compact listing of Trinity based consumer chips:
http://de.wikipedia.org/wiki/AMD_Fusion#Trinity-basierte_Modelle
The english site might be slightly more detailed, but is so needlessly cluttered that I can understand the op...

 

[mrmt]

Right now buying AMD cpus is a lot like buying chinese made generic goods, good luck finding any specs on the product that you took home from the store. Google all you like, there's nothing official on it from the manufacturer.

And this doesn't apply to consumer SKUs only, they give the same treatment for their server and workstation SKUs, and this keep happening on Rory's term at the helm. Says a lot about the culture he is trying to implement at the company.

 

[Charles Kozierok]

Here's a perfect example of how ridiculous it is not having datasheets.

What's the fastest rated DDR3 for use with the Opteron 4300? Well, according to AMD, it's DDR3-1660. Or DDR3-1866. It all depends which document you're looking at.

Does the 4300 have the same IMC as the 4200? It seems to, but I can't tell for sure. It uses the same socket. Did the socket change? I don't know, because I can't access the Socket C32 datasheet.

And so on, with just about everything.

 

[NTMBK]

SDP is a new spec and not applicable to Xeon chips.

Oh, I know it's new and it's obviously not been calculated for older chips. I just making a silly example to demonstrate that Ark is not infallible, although it is clearly far better than the AMD offerings. I'm curious how it is not applicable to Xeon chips, though. Surely a metric intended to indicate power draw under a typical load would be even more useful for server markets which are constrained by performance/watt?