Montecito: 4 Log. CPUs, 667MhzFSB, 24MB cache.....

Mephistopheles · Jul 26, 2004

...the next member of the itanium family, montecito will sport two cores and four logical processors due to the implementation of multithreading (HT-like technology - I wonder if it will utilize Itanium's execution resources better!). Then there'll be speedstep to control power consumption, and there'll be foxton, another new thing about which little details are known. Foxton is supposed to increase speed in the most demanding peaks of operation.... Either through dynamic overclocking or underclocking.

It will also get a 667Mhz FSB (10.6GB/s data transfer) and run at about 2Ghz with 4 times the current amount of cache. Montecito has already taped out...

And after that there'll be Tukwila, with at least 4 cores (possibly up to 16).... Budget versions of tukwila should be socket-compatible with xeon!..... I wonder what they'll do with Itanium until then?...

Without a doubt, montecito will have amazing spec scores... With those gargantuan specifications, it should be a good processor, but will it get better acceptance?...

Will montecito change the current Itanium acceptance? What about Xeon-socket compatibility for tukwila with multiple cores? Sounds theoretically interesting... Itanium has a lot of features and a lot of resources have been pumped into it. Just imagine how a 24MB-cache Xeon with simpler 64-bit extensions (like x86-64) and a 667Mhz, 128bit bus would be like today to get a grasp of the enormous effort that has been directed towards Itanium...

Any comments or more info?...

Edit:Montecito will probably also go with DDR2; dual-channel DDR2-667 should feed its bus happily. Edited by Mephistopheles on 07/25/04 11:19 PM.

Snorkius · Jul 26, 2004

Link...?

The day Microsoft will make something that doesn't suck is the day they'll start making vacuum cleaners.

Mephistopheles · Jul 26, 2004

Here it <A HREF="http://www.vr-zone.com/?i=1021" target="_new">is.</A>

BTW, Montecito is the first chip with more than 1 billion transistors; it'll feature 1.7 billion transistors (~500 million of those are for logic)

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller Edited by Mephistopheles on 07/25/04 11:22 PM.

trooper11 · Jul 26, 2004

well i still think itanium is a dark horse. it seems like it could almost break through if there was that last push, but that hasnt come yet. the way intel is barreling ahead, pumping out out of this world specs and numbers, you have to give intel credit for being brave enough to bet the farm on itanium's long term future. if this will pay off is still uknown, it hasnt yet, this upgrade alone may not be enough.

Mephistopheles · Jul 26, 2004

I think that currently, madison is definitely suspicious... but what I am wondering is if Intel will manage to give it that last push... They certainly seem to be banging their heads against a wall so hard that they might just manage it!!!

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

Kanavit · Jul 26, 2004

holy... hyperthreading on itanium? Intel ain't messin around!!

------
Prescott 3.0E 1MB L2 HT
1GB PC 3200 Dual channel(PAT)
Asus P4P800 Bios 1016
PNY Geforce 6800 GT 256MB DDR3
58,556 Aquamarks

Mephistopheles · Jul 26, 2004

No, I don't think they can afford to just mess around...

Ever thought of it? HT on Itanium is supposed to offer +10% or so performance, plus a 33% increase on clock rates over current generation, and a doubling of cache (per core) over the current generation... Within a year or so...

Then there's silverdale, foxton, pellton, speedstep, and god only knows what they've put inside those 500 million logic transistors... Now only if this thing would be socket-compatible with xeon.... like some tukwila cpus will be...

With this amount of resources, they might as well put in the 30 million logic transistors required for a northwood-like 32-bit processor on that chip... too bad the logic is so different.

The way I see it, either Intel makes Itanium shine like it's the most powerful entity in the universe and make it complement current x86-64 tech in the xeons with extras, or they will have wasted billions in useless R&D... Personally, I hope, for the sake of change and competition, that they make Itanium a monster.

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

P4Man · Jul 26, 2004

I think we'll have to wait for Tukwilla to get a real idea of what IPF can (or can't) do. Montecito doesn't look all that impressive to me..

>HT on Itanium is supposed to offer +10% or so performance,

HyperThreading isn't the word here, its really a different (and vastly less complex) thing. You can't do simultaneous multithreading (like HT) if you can't do OoOE. Montecito will implement coarse grain multithreading, aka switch on event. it basically lets the cpu switch to another thread if a certain thread is stalled for too long (eg waiting for memory). Also, I've seen higher estimates of the performance boost it would bring, but time will tell

>plus a 33% increase on clock rates over current generation,

Which isnt too impressive if you consider IPF has been at 1.5 GHz since last summer, so that is a 30% clock increase in over 2 two years..

>And a doubling of cache (per core) over the current
>generation

More cache is always better, but the further you take it, the more you get diminishing results. Its not like current IPF's are cache starved or something, so I'm not sure this will give another big performance boost. SPECFP will benefit though, it always does when you up the cache

All in all, I doubt Montecito will be enough to recapture the performance crown from Power5(+) which is not good. To establish a completely new ISA, you must demonstrate its advantages over its competitors, and there just aren't (m)any when you compare it to Power. I'd WAG intel/HP will be succesfull in capturing a large part of the (rather small) PA Risc and Alpha market, since those ISA's are discontinued, but I'd WAG they will loose more customers to x86 than they would gain from IBM or Sun. IOW, IPF will IMHO remain a niche product in a shrinking market.

> Personally, I hope, for the sake of change and
> competition, that they make Itanium a monster.

Why would you hope for change for the sake of change ? You are aware switching from one ISA to another costs countless billions to our society ? Fine by me if there is an advantage, but if there isn't, I would hope that money gets used in a more efficient manner.

= The views stated herein are my personal views, and not necessarily the views of my wife. =Edited by P4man on 07/28/04 03:36 PM.

Mephistopheles · Jul 26, 2004

The advantage is pretty obvious, P4Man. When running code that has been written for it, itanium is a very powerful architecture.

You are aware switching from one ISA to another costs countless billions to our society ?

Why does it sound as if you're pulling this number out of thin air? Of course I want there to be an advantage; that's what I said! I wanted it to be bettered and made into something worthy of a change. I also mentioned the compatibility that I2 would get if it were on the same socket as Xeon - which would imply that Itanium would be made better-suited to run the same apps as xeon, or this would have no meaning.

So quit making rhetorical questions (which, BTW, seem almost insulting in nature).

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

P4Man · Jul 26, 2004

>The advantage is pretty obvious, P4Man. When running code
>that has been written for it, itanium is a very powerful
>architecture.

Its no better than existing architectures, so what exactly is the advantage ? More choice might be good, but replacing one ISA with another -incompatible one- that doesn't bring any benefit over the older ones is not good.

>Why does it sound as if you're pulling this number out of >thin air ?

You tell me. Maybe because you do not realize intel alone has poured several billions into IPF ? Add to that the investments made by HP (hardware, software, OS, ..) and "mulitbillion" is already warranted. Add to that investments made by Microsoft, Siebel, SAP, and ~1500 other ISV's, and what do you get ? Add to that the even larger investement required by tens of thousands of enterprise customers that have to buy new hardware, validate it, buy new software, port their applications, validate them, roll them out.. and how would you call it ? Multi million ? If you want a number pulled out of thin air, I'd say $100 Billion over the next decade, but "multi billion" is obviously a given.

Granted, not all of that cost is "extra", and if the industry had stuck with Power, Alpha, x86 etc, a part of that sum would have been spent on hardware purchase, R&D, or software migration and/or development, but not nearly as much.

>So quit making rhetorical questions

As you wish.. next time I'll reformulate that as "It seems you do not realize..", if that makes you happier. sheesh..

= The views stated herein are my personal views, and not necessarily the views of my wife. =

Mephistopheles · Jul 26, 2004

Its no better than existing architectures,...

Isn't Itanium considered a floating-point powerhouse?... It does have some advantages.

You tell me. Maybe because you do not realize intel alone has poured several billions into IPF ? Add to that the investments made by HP (hardware, software, OS, ..) and "mulitbillion" is already warranted.

I do know that, but frankly, I don't care what Intel does with their money. When you said "the society pays", I thought about the sum of average joes, and they're not the ones paying that bill. So maybe I posted a few confusing messages indeed, sorry.

Oh well, whatever, I didn't want to pick a fight, I was merely trying to open my mind to the possibility that, after Montecito's arrival, Itanium will get more interesting and with tukwila's socket-compatibility with xeon, things might change. That's all...

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

trooper11 · Jul 26, 2004

The way I see it, either Intel makes Itanium shine like it's the most powerful entity in the universe and make it complement current x86-64 tech in the xeons with extras

you really think intel wants to compliment x86-64? they dispise it and they push out EM64T kicking and screaming. the transtion to a compatible platform for xeon and intanium looks like to me that intel's long term plan is to phase out the xeon for itaniums across the baord, ending the x86-64 reign. they havent wanted x86-64 and id doubt they would just all of a sudden live with it and let both coexist for long, they already invested this much...

Mephistopheles · Jul 26, 2004

That's another possibility... they might be supporting x86-64 in the form of EM64T, but they might try to sabotage it by replacing it with IA64 after the socket-compatibility takes place. At this point in time, however, they have no choice but to widely support x86-64 or they'll be behind in the eyes of the consumer!!!

In any case, it's really hard to predict everything that will happen... Just try to imagine what'll happen within the next three years... it's impossible to tell...

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

Xeon · Jul 27, 2004

HyperThreading isn't the word here, its really a different (and vastly less complex) thing. You can't do fine grained multithreading (HT) if you can't do OoOE. Montecito will implement coarse grain multithreading, aka switch on event. it basically lets the cpu switch to another thread if a certain thread is stalled for too long (eg waiting for memory). Also, I've seen higher estimates of the performance boost it would bring, but time will tell

How do you know, oh wait you don't, you have no idea how they implemented the technology. But you got the grain thing right I'll take that statement with a grain of salt.

Which isnt too impressive if you consider IPF has been at 1.5 GHz since last summer, so that is a 30% clock increase in over 2 two years..

Well see when AMD gets to 500million transistors and see how easy it is to scale something of that magnitude.

More cache is always better, but the further you take it, the more you get diminishing results. Its not like current IPF's are cache starved or something, so I'm not sure this will give another big performance boost. SPECFP will benefit though, it always does when you up the cache

Cache is cache don't want to little and can't really have too much IMO.

Why would you hope for change for the sake of change ? You are aware switching from one ISA to another costs countless billions to our society ? Fine by me if there is an advantage, but if there isn't, I would hope that money gets used in a more efficient manner.

WTF are you some pot smoking liberal or something "cost countless billions to our society. What it god's green bloody frikken world does that have to do with anything, that’s Intel's money we paid into it and they are using it.

It's like saying your boss has a say over what you spend your money on. Stupidest [-peep-] I have heard all month man stupid stupid and more stupid.

Its no better than existing architectures, so what exactly is the advantage ? More choice might be good, but replacing one ISA with another -incompatible one- that doesn't bring any benefit over the older ones is not good.

Hmm well since you don't know I do, there is lot's of information out there for you to read <A HREF="http://www.extremetech.com/article2/0,1558,1155598,00.asp" target="_new">this</A> even though not pro Intel does mention some of the better points hence why I chose it.

You tell me. Maybe because you do not realize intel alone has poured several billions into IPF ? Add to that the investments made by HP (hardware, software, OS, ..) and "mulitbillion" is already warranted. Add to that investments made by Microsoft, Siebel, SAP, and ~1500 other ISV's, and what do you get ? Add to that the even larger investement required by tens of thousands of enterprise customers that have to buy new hardware, validate it, buy new software, port their applications, validate them, roll them out.. and how would you call it ? Multi million ? If you want a number pulled out of thin air, I'd say $100 Billion over the next decade, but "multi billion" is obviously a given.

Granted, not all of that cost is "extra", and if the industry had stuck with Power, Alpha, x86 etc, a part of that sum would have been spent on hardware purchase, R&D, or software migration and/or development, but not nearly as much.

Again I will ask why it is any of our concern what Intel corp. does with there bloody money. You should start riding AMD those fools waste more money than any other semiconductor company in known history.

Xeon

Post created with being a dickhead in mind.
For all emotional and slanderous statements contact THG for all law suits.

P4Man · Jul 27, 2004

>How do you know, oh wait you don't

Unlike you, I tend to know at least a bit of what I am talking about. I am not speculating, I am telling you.
1) Montecito will implement course grain multithreading (aka switch on event, not SMT like P4's hyperthreading
2) You can't possibly do simultaneous multithreading without OoOE logic.

If you don't believe me, I don't care, but do so some reading. Im not telling anything that isnt widely known and published all over the web.

>WTF are you some pot smoking liberal or something "cost
>countless billions to our society. What it god's green
>bloody frikken world does that have to do with anything,
>that’s Intel's money we paid into it and they are using it

Its not just intels money; its the money of its customers, partners and shareholders as well. I don't blame intel for making or spending money, not for wasting it (I'll let shareholders complain over that one), but its my opinion that its a pretty spectacular waste of money for the entire industry, (and therefore economy and society) if 3 platforms are killed and replaced with a new one if the only reason is increasing intels profits. So far, the IPF ISA has not demonstrated any inherent advantage over Alpha or Power, so its a damn shame those billions are being spent everywhere just so that perhaps intel could profit one day. Its like if some major electricity providers would change from 110V to 130V, (assuming there would not be any benefit) thereby requiring everyone to change appliances, cabling, wall sockets for no other reason than their own benefit (assuming there would be one). Not my idea of progress, but then i'm not an intel shareholder..

= The views stated herein are my personal views, and not necessarily the views of my wife. =

P4Man · Jul 27, 2004

> I was merely trying to open my mind to the possibility
>that, after Montecito's arrival, Itanium will get more
>interesting and with tukwila's socket-compatibility with
>xeon, things might change.

Oh they will change, that much is for sure. Once customers can't buy PA Risc of Alpha anymore, IPF as their natural successor will obviously capture the majority of those markets. And since no futher development will be done on PA Risc or Alpha, IPF will obviously outperform them (not sure if you realize, but even today IPF has a hard time beating the ~8 year old design and 180nm based Alpha products!)

But wether or not IPF will be able to really outpace Power (and even x86) like Alpha once did, and thereby demonstrate why the world needed EPIC over RISC, IA64 over alpha, I remain very sceptical. Perhaps with Tukwilla though.. too little is known to make any sort of guestimate.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

trooper11 · Jul 28, 2004

Well see when AMD gets to 500million transistors and see how easy it is to scale something of that magnitude.

lets hope that is years down the road lol. amd would be crazy to try and develop something like that in the scale of itanium. intel cna afford to loose if something doenst work out. amd is almost always risking its entire company on its new product line.

i dont believe that more cache or more transistors is a good thing, its one thing to add more and at the same time making use of those things mroe efficently, but i just odnt think the future is in making huge sizes of transistors and adding piles apon piles of cache. how about someone focus on refining and making the current sizes mroe efficient? why is that so hard to understand. at some point they will hit a wall they cant get by and forced to re think thier strategy.

its just like the mhz race, everyone figured out its not the ammount, its how you use each tiny bit. amd had this idea form the start, and intle picked up on that. but it seems they think it only applies there, but it doesnt. less is better, as far as amount goes.

i also think that saying a chip is the best using proprietary software is a slippery slope, since any chip can be faster on software designed specifically for it, take amd64 for a an example.

so if intel can force all software developers to make software that boosts its products and no others, then good for them, although that would be illegal lol. there is no standardization in software, you have some apps that are nuetral and work fine on all platforms, giving edges to one or the other sometimes if there is a real difference between the platforms. then there are those that will highly favor one over the other. and this goes on in every segment. for instance, every app someoen can show that favors intel for audio encoding, someone else could show apps that favor amd. this is why software benching is a joke, youll never get everyone to agree to one standard of creating apps, and you know amd and intel would raise hell if one way was decided apon.

P4Man · Jul 28, 2004

I have to defend the Itanium strategy here (oh irony)..

> dont believe that more cache or more transistors is a good >thing,

Of course it is.

>its one thing to add more and at the same time making use
>of those things mroe efficently, but i just odnt think the
>future is in making huge sizes of transistors and adding
>piles apon piles of cache. how about someone focus on
>refining and making the current sizes mroe efficient?

Well, over the last 20 years heroic efforts have been taken to achieve just that, and with some spectacular results I might add.. but now we seem to be getting to point where CPU cores are about as "clever" as they will get, and since they are getting smaller as well (in spite of exponential increases in transistor count), it might make more sense to add more cores than to further only try and enhance the cores. These cores are getting so ridiculously complex that is just too frigging hard to design or debug them, or simulate them. Its not a coincidence you see all major MPU manufacturer moving towards multicore (Intel, AMD, but also IBM (Power for a long time), Sun (niagra anyone ?), Sony/IBM (Cell), even HP made its last PA Risc chip dual cored.

>at some point they will hit a wall they cant get by and
>forced to re think thier strategy.

I think its the opposite.. Increasing IPC (ILP/TLP) will only get you so far. Sure there is some headroom left, and clockscaling isnt entirely dead yet either, but the low hanging fruit is mostly gone especially if you'd combine the "best of all worlds" today like, ODMC, Hypertransport but then with HT, tracecache, etc,.

If there is a brick wall, its IMHO more in the IPC race than in the transistor count race. What else are you going to do with 65nm or 45nm devices than putting 4-16x as many cores on there ? And cache is pretty damn cheap as well.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

Xeon · Jul 28, 2004

If you don't believe me, I don't care, but do so some reading. Im not telling anything that isnt widely known and published all over the web.

Your right I don't give a flying crap what you have to say on this topic. Your ego out paces your views as far as I am concerned. You have no links, no technical back ground, and your European.

At least when I argue with Fugger, Mephitsopheles, Trooper, Cibercraig, TheRod, Spitfire, Crashman, Phial even thegreatgrapeape. They all provide links that are valid or they win me over with a well thought argument.

You just go hey, believe me I'm from Europe.

but then i'm not an intel shareholder

The stop whining about a American company that you don't purchase from, like, need, or own any part of. When you run a multi billion dollar company your opinion will matter on the field of large corporations spending large amounts of money to replace a ancient and clumsy IA like x86.

Ill say it again, it’s like saying your boss has a say over what you spend your money on. Stupidest [-peep-] I have heard all month man stupid stupid and more stupid.

too little is known to make any sort of guestimate.

Wish that rule applied to most of the crap that comes out of your mouth.

If there is a brick wall, its IMHO more in the IPC race than in the transistor count race. What else are you going to do with 65nm or 45nm devices than putting 4-16x as many cores on there ? And cache is pretty damn cheap as well.

Hmm like AMD will make it past 0.65u, and we all know the costs that are associated with those next two shifts. But the IPC race will never be over. There are engineers that live on the idea of taking silicon logic and making it more efficient and smarter, just as there are engineers that live off of making transistors switch faster and scale higher.

Xeon

Post created with being a dickhead in mind.
For all emotional and slanderous statements contact THG for all law suits.

P4Man · Jul 28, 2004

That's a pretty pathetic post. Go dig up your own links if you want proof, you ain't my judge, and I ain't your secretary. As for ego, weren't you the guy that copy pasted Intel whitepapers as claimed them to be your own work ? Sheesh...

I have no further interest in continuing this and I'll let others make up their minds if they want to.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

trooper11 · Jul 28, 2004

well i dont know, it just seems to me that if we are at the point that the core cannot be tuned any further, which i dont belive is true at all, just consider carbon nanotubes and the like, that just piling up cache is not the answer. even if your true and the core is as far as it can go, that would be prettty sad if you ask me, that all companies now have to resort to pumping up cache levels and adding more and more transistors. transistors will fade into the sunset eventually, left behind by nanotubes and other forms of nano tech ,w hcih wil enable great strides in cpu core development. i think what you see today is the limits of the current tech, ill grant you that, but i dont think that becuase of that, comapnies should just decide to pile things onto an aging tech. if anything, id be investing more into future tech that will get us past these stumbling blocks.

P4Man · Jul 28, 2004

>even if your true and the core is as far as it can go,

Thats not what I said or wanted to say; just that it gets exponentially more difficult and the IPC increases ever smaller. I've been looking for some benches to put numbers on this, but I can't find anything that gives me scores from ~486 to up to the current P4's to show my point (if anyone knows any, please post/link, I'd be most grateful). I'm fairly certain performance per clock as well as performance per transistor is (considerably) lower now than back then. Performance per $ spent on R&D would be even more interesting, but we won't get those numbers I fear

I just see a trend; single threaded performance seems to be levelling off slowly (both IPC and clockspeed) while fab capacity and transistor budgets (per die) continue to "explode". Again, I'm not claiming we have hit the IPC or clockspeed wall at all, and there most likely isnt one, but if you'd graph IPC and clockspeed over time, I'm fairly sure you'd see a slowdown that would not show up (or not as pronounced) when you'd graph transistor budgets, and definately not when you'd chart R&D costs. Therefore, I'd argue spending the extra transistor budgets on cache and multiple cores makes a lot of sense.

Mind you, IMO single threaded performance is still extremely important, and will remain so for the foreseeable future (at least on the desktop), but it is much harder to achieve than multithreaded performance (ironically, the opposite applies to software). there will be a point, not too far from here, where it really hardly matters wether you put 1, 2 or even more cores on a single die from a cost perspective. Getting similar potential performance increases in single threaded performance OTOH would be incredibly expensive, if not downright impossible.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

Mephistopheles · Jul 29, 2004

I'd say that, right now, it's much easier to design multiple cores than develop better cores, so that explains taking this direction. It doesn't mean there's no way to manufacture a better core, it's just that, utilizing current tools and current paradigms, it's just too damned expensive. I agree with P4Man in that regard.

You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete - Buckminster Fuller

trooper11 · Jul 29, 2004

well its been pretty clear taht loading up on cache has a limited performance benefit, it tends to level off at some point. for instnace, you dont see intel ever drop back its levle of cache from one core to the next, its always aiming for higher. but you see amd cut cahce in half when moving from clawhammer to what is the s939 cores of today and its no worse for the ware. all im saying is cache isnt the answer.

as far as multiple cores goes, i think tis a great idea too, its a great way to cheat the system so to spek and give a breath of life to the aging tech. this gives developers time to invest into new forms of redesining cores. i think the excuse that more should not be invest simply becuase its too expensive is bogus, we are talking about an industry where the end can be seen with the current tech, the whole reason we are moving to dual core should make that obvious. eventaully multiple cores will become to difficult to improve apon. good lord, why would it be so bad for intel especially, they are willing to spend billions on itanium and they have these huge R&D facilities, to devote resources to developing these new transistors. I have a feeling they are already,t ehy would be stupid not to be doing so. We all know IBM is. quantum computing is getting closer and closer, albeit stil many years away. im just saying i think we have htia wall and the measures taken now are mearly stpes to keep it on life support till the tech and economics of new tech comes into range.

i mean look at how close mad and intel are in performance in many areas, its getting harder and harder to make clear seperations, so you see the focus shift from performance to features.

Johanthegnarler · Jul 29, 2004

You could be a war mongering rebuplican that doesn't really do anything positive for our country.. or a democratic lush that just talks so the people vote. The way i look at it.. both parties have made our country look pathetic in the eyes of the world. We are a losing civilization and will be taken out as history repeats itself.

As for itanium, they can stall it another 5 years and not have a problem. There's no real need for it yet. Software is just now getting to the 64 bit world and that's even slow. I have a feeling software is going to cap off for awhile before it even makes descent use of 64 bit apps.

<A HREF="http://arc.aquamark3.com/arc/arc_view.php?run=277124623" target="_new">http://arc.aquamark3.com/arc/arc_view.php?run=277124623</A>
46,510 , movin on up. 48k new goal. Maybe not.. :/

Montecito: 4 Log. CPUs, 667MhzFSB, 24MB cache.....

Distinguished

Splendid

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Share this page