Unstoppable: DDR400 vs. Rambus

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BDSM

Senior member
Jun 6, 2001
584
0
0
fkloster.. You're wrong it does have a configuration, overclocked.

Intel MAY release a 32 bit pc1200 utilizing chipset sometime in the future. Let's wait for the 16 bit pc1066 one that hasn't been released or even announced yet and then see what happens.

Dullard.. hehe. well.. we could go on doubling forever. A quad Hammer system will have a total bandwith of 4.2*4=16.8 GB/sec.
But I feel it's pointless going that far into the future. Dual channel DDR is here and has been for some time now. There is the N-force and there is now the new i7500 (dunno the exact name of that one).

Btw... are the forums really slow today or is it just me?
 

fkloster

Diamond Member
Dec 16, 1999
4,171
0
0


<< fkloster.. You're wrong it does have a configuration, overclocked. >>



Please provide a link to ANYONE overclocking there PC-800 or PC-1066 to PC-1200.....I have not seen this yet...
 

fkloster

Diamond Member
Dec 16, 1999
4,171
0
0


<< Dual channel DDR is here and has been for some time now. There is the N-force ... >>



lol & we all know how well Dual-Channel nforce compares to Dual-Channel PC-800 systems
 

fkloster

Diamond Member
Dec 16, 1999
4,171
0
0
Aw thanks I found it...




<< In this test, we use hand-picked memory modules ranging from DDR400 to RDRAM (PC100). The results show that while DDR200
lands at the bottom of the ranks, RDRAM with 533 MHz and 600 MHz achieve the highest performance levels.
>>

 

dullard

Elite Member
May 21, 2001
25,886
4,469
126
"Intel MAY release a 32 bit pc1200 utilizing chipset sometime in the future."
I agree PC1200 is too far in the future to consider.

"Let's wait for the 16 bit pc1066 one that hasn't been released or even announced yet and then see what happens."
Information on it is leaked in several Intel webpages (including the one I posted earlier on this thread). It should be released early May, so for all this discussion it is safe to say it is here.

"we could go on doubling forever...Dual channel DDR is here and has been for some time now. There is the N-force and there is now the new i7500 (dunno the exact name of that one)."
N-force has dual channel DDR but the processor cannot access all the bandwidth (correct me if I'm wrong, but one channel is for the video card and the other for the proccessor - leading to the poor performance we have seen so far from it.) The i7500 for Intel Xeons is here and does seem to top PC800 in the few benchmarks I've seen. However the lead was slight since the i7500 uses slow DDR so PC1066 will easily surpass the i7500 (I do not know if there will be PC1066 for the Intel Xeon very soon). I still haven't seen any dual channel DDR that can work with fast DDR and has both channels accessable by the processor.

"Btw... are the forums really slow today or is it just me?"
This thread is a slow loader for me, it might be due to the long posts people have written. The rest of Anandtech seems about the same to me.
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76


<< Intel MAY release a 32 bit pc1200 utilizing chipset sometime in the future. Let's wait for the 16 bit pc1066 one that hasn't been released or even announced yet and then see what happens. >>

True, there's nothing certain onRDRAM's future beyond 16-bit PC1066P

<< Yes, THG showed you the numbers. He shouldn't have. He could just as well have put on the numbers for DDR 2.. or 3.
Or DDR 87 or something and you would have bought it just as well.
>>

What's that supposed to mean? I looked at the numbersd and testing platform and made the conclusions myself. Please tell me what I am "buying" that isn't true.

<< Any 3D app with a directx8 engine is going to perform better with DDR than PC800 Rambus, period. This is a general statement, dependent on several other factors, but it is more or less true. >>

I'm sorry, but where on earth did you get that? Please give me a link that proves this.

<< I've noticed that many of you in this thread can't get over quoting these ridiculously high memory bandwidth scores of RDRAM (9 Gb/s - so what?). You are completely ignoring the fact that high overall bandwidth means very little compared with memory latency since only a fraction of total DDR or RDRAM bandwidth is currently used by 3D games. It is going to be years before any game needs more than 2GB/s per second memory bandwidth across the graphics bus, and processor to memory transfers don't necessitate more than that amount either. The raw numbers that benchmarks spit out for performance on P4+1066 mean squat because no apps are using all of that bandwidth. Memory latency comes into play much more for todays 3D apps, and in that department DDR has RDRAM beaten. It is really that simple.
>>

I have yet to quote any Sandra Numbers, and my conclusions about PC1066 come from its real world performance in things like new current games, MP3 and MPEG4 encoding and the like.

<< Memory latency comes into play much more for todays 3D apps, and in that department DDR has RDRAM beaten. It is really that simple.
>>

Again, numbers please.

<< Correct - we don't need much more bandwidth at the moment. However along with the higher bandwidth comes significantly lower latency and an increased fsb. These give a significant performance gain. Just look at the 14% faster games when going from PC800 to PC1066. Or 24% faster games when going to PC1200. Neither result is 'a futilly small gain' in my opinion. It isn't the bandwidth giving the boost, it is the lower latency and increased fsb. >>

I agree. Really, as long as the fsb increases, there's nothing stopping PC1200 from improving alot.
 

Zakule

Member
May 1, 2002
35
0
0
The answer to all three of your questions is yes, Fkloster. 3DMark2001 is Directx based, RDRAM does yield higher scores, and everything I've posted here is true to the best of my knowledge, however I'm finding myself ill-equipped to deal with your ignorance.

3DMark2001 includes AGP testing, T&L test, and a few other features that are more or less not used in most games. Game developers currently code their games in a much different fashion than 3DMark2001 and although I could explain to you exactly what the differences are I doubt you'd listen since I'm getting the impression you've long since decided that you know everything about how computers work.

Here's a general description for you to digest. 3D textures are rendered off of video memory not system memory so AGP 4x doesn't really mean squat and yet 3DMark2001 uses AGP 4x in some of it's benching. Games only use AGP to move data to the video card for rendering because rendering with the AGP system via the AGP aperture takes forever by comparison. T&L in games is almost non-existent but is a large portion of 3DMark2001's testing. Ergo, 3DMark2001's number mean very little.

Keep going, you're getting closer
 

BDSM

Senior member
Jun 6, 2001
584
0
0
kloster.. how convenient that you "forgot" to quote me corecctly, you left out the e7500 chipset which performs very well compared to the i860 coupled with RDRAM.. and that is with DDR200!!
It will compare VERY nicely to the i860, coupled with DDR266. And no matter what some say DDR266 IS cheaper than RDRAM and always will be.

The Athlon can't make use of the 4.2 GB/sec that the N-force offers but the p4 can @ 133 mhz bus (533 mhz data rate).

I was merely proving to you that dual channel ddr doesn't cost that much.. it's a myth.. Just like the supposedly cheap RDRAM based four layer p4 mobo's we were going to see everywhere.. but where are these mobo's??

Instead athlon DDR boards cost alot less.. How is that?!?!
 

fkloster

Diamond Member
Dec 16, 1999
4,171
0
0
Gotta go Zakule. I will be back early morning tomorrow...please don't go away...my friends and I are not even close to being done with you yet
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76


<< The Athlon can't make use of the 4.2 GB/sec that the N-force offers but the p4 can @ 133 mhz bus (533 mhz data rate). >>

I agree totally.

<< I was merely proving to you that dual channel ddr doesn't cost that much.. it's a myth.. Just like the supposedly cheap RDRAM based four layer p4 mobo's we were going to see everywhere.. but where are these mobo's?? >>

Maybe, I'm just not convinced that Intel's solution will be priced low enough.

<< The answer to all three of your questions is yes, Fkloster. 3DMark2001 is Directx based, RDRAM does yield higher scores, and everything I've posted here is true to the best of my knowledge, however I'm finding myself ill-equipped to deal with your ignorance.

3DMark2001 includes AGP testing, T&L test, and a few other features that are more or less not used in most games. Game developers currently code their games in a much different fashion than 3DMark2001 and although I could explain to you exactly what the differences are I doubt you'd listen since I'm getting the impression you've long since decided that you know everything about how computers work.

Here's a general description for you to digest. 3D textures are rendered off of video memory not system memory so AGP 4x doesn't really mean squat and yet 3DMark2001 uses AGP 4x in some of it's benching. Games only use AGP to move data to the video card for rendering because rendering with the AGP system via the AGP aperture takes forever by comparison. T&L in games is almost non-existent but is a large portion of 3DMark2001's testing. Ergo, 3DMark2001's number mean very little.
>>

I admit that I do not have the knowledge about these things, so I'll take your word for it.
 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
Oh wow... this has started to turn into a flame war...

RDram has on average higher latency than the SDram varieties. However, when the bandwidth is stressed, RDram will have *less* latency than any SDram variety. That is why RDram can easily reach its peak bandwidth much more easier than its SDram brothers. Thats why DDR400 SDram on an i845/Sis645DX platform BARELY manages to squeak by a PC800 system. Because of this, RDram will actually be better at high memory demanding applications (3d, compression, video encoding/decoding, blah blah) while it will comparatively suck at normal applications.

As for E7500 vs i860. You must think this out. E7500 is only operating at Dual channel DDR200 because of hardware limitations (Intel had problems with the clock multiplier to the frontside bus). You must also realized that e7500 is top of the line technology wise and that the i860 is almost 2 year old technology! Besides, RDram is hardly meant for serving. The more modules you add to it, the more latency that it will have (damn serial connections). RDram, IMO, shold only be used for high performance desktops/workstations.

Extremetech reviewed PC1066 RDRAM back in November of 2001. It compared it to DDR333 P4 setups, as well as AthlonXP machines. Their benchmarks showed PC1066's superiority for app's like Quake 3, but it also pointed out it's weaknesses and briefly discussed latency issues (I think, it's been a while since I read the review).

Yes thats is *very* true. However, that was November 2001, willamettes were barely pushing 2Ghz. Now that Northwoods are pushing over 3Ghz with an improved core, the performance gap of PC1066 and DDR SDram has intensified because of the CPU that is ABLE to demand the bandwidth offered by RDram. Before that time, the RDram bandwidth was more or less useless to the Pentium IV. With the advent of higher clock speeds, there comes the need for MORE memory bandwidth, especially on CPU intensive and memory intensive applications. With PC1066 RDram, the P4 enjoys a significant lead across the board against non-RDRam p4s, and that lead will only GROW as p4s become faster.

As for pricing, RDram is *cheaper* than DDR SDram. the best RDram (Samsung PC800) can be had for $70-80 for a 256 stick. Comparatively, a DDR266 CL2 from crucial is ... last time i checked, $90. Yes DDR SDram can be cheap, but those are either non-name brands OR BAD LATENCY (non CL2). With the cost of the motherboard included, the price is evened out.

Yes, DDR2 is still in development and I am fortunate to have gotten a sneak peek at what it can do. That's all I can and will say about it, and again I can't give you link to a review site, which are seemingly your only sources of information - how unfortunate for you.

Correct me if im wrong, but the only DDR2 chipset i have seen is a McKinley based server. Which btw, is a rather interesting piece of technology... but enough said about that. I dont work for AMD, so I cannot comment on the Opteron.

As for tomshardware and Tom pabst. You do realized that Tom Pabst was the #1 Rambus hater on cyberspace when the i820 and i840 chipsets for the Pentium III was out. Frankly, I find his reviews to be biased in his opinions, but his numbers usually dont lie. Even though he is biased, I do think he knows much mroe about computer hardware than Anand (tom can also be a bit pyscho too, imo).
 

Zakule

Member
May 1, 2002
35
0
0


<< I admit that I do not have the knowledge about these things, so I'll take your word for it. >>



No need to take my word on it, Athlon4all. Take a trip over to viahardware forums and ask some of the more experienced users there.

Everyone gets so excited about more bandwidth across the AGP bus when the bottleneck is actually on the video card itself. As of AGP2x we've got more than enough AGP bandwidth then is actually used. AGP4x is just ridiculous since even the most graphically instensive app won't even use on third of its available bandwidth.

AGP was pretty much dead on arrival but because of Intel's influence in the market we were forced into using it. The original idea behind AGP was to save costs on memory (which was very expensive at the time of AGP's inception) by making system memory accessible by the video card as local memory. That way you wouldn't have to pack a bunch of memory onto the video card since the video card could just use system memory set aside for it specifically called the "AGP aperture". As 3D accelerator technology progressed and memory began to drop in price AGP became an obselete system before it was even released. Main reason - the all powerful GPU. That's right, as soon as you move the bulk of the rendering ops to a video card with it's own processor and fast memory then the AGP system of rendering looks pretty damn slow.

Currently, games are coded to cache their textures on the video card and let the video card render frames. The amount of data that actually moves across the AGP bus while the video card is spitting out frames is usually less than 200 MB/s. You see the video card already has everything it needs except coordinates on where it is to paint the textures.

If anybody ever wrote a game that used AGP for 3D rendering nobody would buy it because it would look terrible. Imagine having to wait for the processor to send the texture to memory and then to the graphics bus before it gets rendered. Why would anybody do that when the GPU+DDR on the video card can make the same calculations and render the frame in about one tenth the time it would take for the system processor to accomplish the same task with main memory? You wouldn't, and currently no game designers do.




<< Gotta go Zakule. I will be back early morning tomorrow...please don't go away...my friends and I are not even close to being done with you yet >>



Ah, I guess 6th period just ended, huh! :Q

See you tomorrow, my overzealous, under-educated adversary!
(sorry about the flame, but I just couldn't help myself. You are too easy )
 

BDSM

Senior member
Jun 6, 2001
584
0
0
devxvx.. that's the thing with Tom.. he "hated" rambus... that somehow changed.. now he hates other stuff.. I wonder what it is that makes him change the object of his hatred back and forth?
 

Zakule

Member
May 1, 2002
35
0
0


<< devxvx.. that's the thing with Tom.. he "hated" rambus... that somehow changed.. now he hates other stuff.. I wonder what it is that makes him change the object of his hatred back and forth? >>



I can answer that question. Tom doesn't know what the hell he's doing. He knows just enough to sound smart and that's about it. Many, many of his benchmarks have been flawed and I think Tom's Hardware Guide is much more of a detriment to the tech community the it is an asset.

I can still recall his "shadow of Rambus doomed to fail" article and now, I guess, he's okay with it. What a putz.
 

Athlon4all

Diamond Member
Jun 18, 2001
5,416
0
76


<< devxvx.. that's the thing with Tom.. he "hated" rambus... that somehow changed.. now he hates other stuff.. I wonder what it is that makes him change the object of his hatred back and forth? >>

Well, I think a key thing that must be kept in mind is that Tom isn't doing most of the articles on THG any more sadly, that's part why I feel AT is a more reputable site than THG, is the guys replacing Tom aren't as good IMHO, but anyway back to topic:

<< Everyone gets so excited about more bandwidth across the AGP bus when the bottleneck is actually on the video card itself. As of AGP2x we've got more than enough AGP bandwidth then is actually used. AGP4x is just ridiculous since even the most graphically instensive app won't even use on third of its available bandwidth.

AGP was pretty much dead on arrival but because of Intel's influence in the market we were forced into using it. The original idea behind AGP was to save costs on memory (which was very expensive at the time of AGP's inception) by making system memory accessible by the video card as local memory. That way you wouldn't have to pack a bunch of memory onto the video card since the video card could just use system memory set aside for it specifically called the "AGP aperture". As 3D accelerator technology progressed and memory began to drop in price AGP became an obselete system before it was even released. Main reason - the all powerful GPU. That's right, as soon as you move the bulk of the rendering ops to a video card with it's own processor and fast memory then the AGP system of rendering looks pretty damn slow.

Currently, games are coded to cache their textures on the video card and let the video card render frames. The amount of data that actually moves across the AGP bus while the video card is spitting out frames is usually less than 200 MB/s. You see the video card already has everything it needs except coordinates on where it is to paint the textures.
>>

I know that already, that AGP 4x doesnt help at all, but to be honest, I fail to see what this has to do with RDRAM vs. DDR400. The reality is that despite much of the rendering is done on chip, still the CPU/Memory are still very much involved (why else would a GF3 and up perform better on a faster CPU?). But anyway
 

BDSM

Senior member
Jun 6, 2001
584
0
0
Zakule.. I belive you are giving him too much credit. I think he's not that incompetent. It's just that his judgement is.. "clouded".. somehow.. if you know what I'm talking baout.
 

Zakule

Member
May 1, 2002
35
0
0


<< I know that already, that AGP 4x doesnt help at all, but to be honest, I fail to see what this has to do with RDRAM vs. DDR400. The reality is that despite much of the rendering is done on chip, still the CPU/Memory are still very much involved (why else would a GF3 and up perform better on a faster CPU?). >>



It doesn't have anything to do with DDR vs. RDRAM, per say, but it has to do with why 3DMark2001 is a half-ass benchmark. It uses features in it's testing that are almost never used in real world apps. Don't blame me for the tangent, Fkloster through it out to refute my claims and I therefore felt it necessary to dispell some of the more common misconceptions about 3DMark2001 and 3D graphics rendering in general.

The system CPU is still a large part of the equation, yes, because it needs to send the video card the necessary data to render frames. Even with all of the textures cached on the video card it still needs to know where to paint them. This is at the heart of why memory latency is so much more important than bandwidth. It doesn't need that much bandwidth but it does need to get the data to the video card ASAP after it's requested.
 

dexvx

Diamond Member
Feb 2, 2000
3,899
0
0
i believe what you are saying is that all review sites are biased in some direction, and the only way to find out whats best is to actually test the situation out yourself. I myself have an Asus p4t-e board with 512MB RDram at 1120. I have ordered an Asus p4b-266C with 512MB DDR 333 SDram CL2 which I will build for a friend. Leaving everything else constant, I still believe that the RDram system will outperform the DDR SDram system by a wide margin, even if the DDR SDram system has a better clock speed.

I for one, don't find tom's benchmarks 'flawed,' by looking at his system setup. In fact, I get similar performance to his test systems (The system im typing on his a similar configuration). I won't go into the pysche of tom, because I find little relevance in that. But i do find it odd that when Tom says RDram is on top, you go and immediately declare he has "changed his object of hatred." He might have, but you are more or less basing it on circumstantial evidence. It seems to me that whenever some 'review site' goes against to what you are saying, you immediately discredit it claiming bias. Excuse me if im wrong, but I believe that is what bias is about.

Getting on subject... you must remember that Rambus's latency rating is for the worst possible scenario and SDram varients are at the best possible scenario. As stated previously, the higher the bandwidth demand, the worse the latency for SDram variants and the comparative advantage goes to RDram. I believe your comparison of RDram latency rating and that of SDram variants are flawed. Unless you really don't trust cachemen, or should I write a program myself that will test and plot the latency versus bandwidth used graph.
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
so what youre trying to tell me is, agp4x is more than enough for the latest card, switch your mobo to 2x and run 3dm2k1, and because 3dm2k1 is so corrupt and isnt "real world" run game #2 (lobby) at high quality 1600x1200x32, game #2 uses the engine for the game Max Payne, a real game. (dragothic is from theif 2 engine) You will see a MASSIVE loss in performance because the card cannot send the data to the cpu or memory. (this will only happen on Geforce3/4/Radeon8500 to extremes)

NVIDIA and Intel are already deving some new slot for the "eclipse" chip coming this fall, because agp 8x isnt a fast enough interface.

And before another flame war ensues, we are talking 130fps instead of 100fps, big frigging deal, buy whatever you want, you wont tell the difference lol.
 

sanz

Member
Apr 23, 2001
160
0
0
Interesting arguments.. Product loyalty? Childish. Product bashing? Unless you've used it, don't bash. Remember... at the end of the day, you buy what you want so try not to force your views on others.

Anyway, back to the topic.. Toms article..

Just another article to reinforce what most on this forum knows already. RDRAM still the best solution for P4. I understand that Tom's hardware is biased.. but why do they keep making themselves sound stupid? Nice article.. then comes conclusion.. Never ceases to amaze me...There must be a reason for that...

Article Title..
Full Speed Ahead: DDR400 vs. RDRAM (PC1200) - DDR400 vs RDRAM. Can DDR compete against RDRAM?

Conclusion
Conclusion: DDR400 Outperforms RDRAM (PC800) - Much Depends On The Chipset

Ah yes.... I can see the logic behind that..

"Lets test DDR400 vs RDRAM (PC1200) and show everyone that RDRAM (PC1200) will be one fast RAM (which everyone knows anyway) and conclude with comparing the lastest DDR400 with 2 yearsish old RDRAM (PC800). Nevermind that it was supposed to be DDR400 vs RDRAM PC1200. We have to make DDR400 look better!"

 

kgraeme

Diamond Member
Sep 5, 2000
3,536
0
0


<<
Ah, I guess 6th period just ended, huh! :Q

See you tomorrow, my overzealous, under-educated adversary!
(sorry about the flame, but I just couldn't help myself. You are too easy )
>>



You complain about petty forum fighting and you are resorting to name calling. Have you considered the possibility that fkloster might work for, say, a memory technology company? Just a thought. Polite discussion might merit more learning.

I personally like DDR, but admit that RDRAM currently wins the benchmark war. I still won't buy RDRAM for socio-political reasons, but that's another matter entirely.
 

kgraeme

Diamond Member
Sep 5, 2000
3,536
0
0


<<
"Lets test DDR400 vs RDRAM (PC1200) and show everyone that RDRAM (PC1200) will be one fast RAM (which everyone knows anyway) and conclude with comparing the lastest DDR400 with 2 yearsish old RDRAM (PC800). Nevermind that it was supposed to be DDR400 vs RDRAM PC1200. We have to make DDR400 look better!"
>>



I too was pretty confused by the entire article.
 

BDSM

Senior member
Jun 6, 2001
584
0
0
Sanz.. You can order a SIS 645 DX mobo and a DDR 400 module and use them together (even though they are still rare), but you can't buy any pc1066 modules and absolutely no pc1200 rdram modules yet and there is no mobo support for neither of them.

The 645 DX at least has some support for 200 mhz bus. Using i850 and pc800 you would have to overclock both the mobo and the memory.

Maybe this is why they draw those conclusions.
 

MadRat

Lifer
Oct 14, 1999
11,965
279
126
<<Someone talked about the "high cost" of dual channel ddr. I don't believe that. Mobo's based on the n-force dual DDR266 chipset can be had for less than $130 here in Sweden. While the cheapest RDRAM mobo I can find is a good $40 more expensive!
>>

The nForce is NOT a dual-channel (i.e. 128-bit pathway) chipset. The nForce is an INTERLEAVED (64-bit) chipset. Interleaving does not improve latency. There is alot of misinformation going on by chipset makers.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |