Question B550 chipset, so AMD joins the dark side after all.

Page 9 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
I just read the article...







So let me get this straight, this chipset is coming out like a year later, they did not even bother to add CPU PCI-E 4.0 uplink support or to increase the number of sata ports that is ALREADY a problem on every 6 sata B450 motherboard (NVME x4 disables the 2 SOC Sata, thus 6 sata B450 mbs losses 2 sata if NVME is used), and they even dare to futher reduce backguard compatibility?

I was not expecting for the PCI-E lanes FROM the chipset to be 4.0, but only USB 3.2 G2, no more satas, CPU link still 3.0 and the PCI-E lanes 3.0 is beyond disappointing.
 
Last edited:
Reactions: PingSpike

zir_blazer

Golden Member
Jun 6, 2013
1,184
459
136
I want to point out that in the AM4 platform, the Chipset is merely a glorified PCIe Switch, SATA Controller and USB Controller. The Processor itself fulfills all the core roles as Zen was designed as a potentially standalone SoC. As such, the Chipset is an extra, thus it doesn't has to support the Processor at all, making it less important than what people thinks that it is. If anything, it is technically possible that Zen 3 works in any Chipset, or maybe even without (A la A300/X300, which were supposed to be just security chips wired to the Processor via simple SPI and not full fledged Chipsets wired via 4 PCIe Lanes).

What you have to focus here is in OFFICIAL SUPPORT. We know that AMD already said no, but that doesn't means that it will not work. If we go for that logic, AMD officially does NOT support Zen 2 on B350/X370 Chipsets, yet they work regardless. Since there are no technical reasons for it to not work, then I expect that it should be possible to get Zen 3 working. Whenever the Motherboard manufacturers decides to support it on their own, or whenever you require some BIOS modding to force newer AGESAs onto older Motherboard (As if that wasn't currently the case...), is something still yet to be seen and there is nothing else that you can do but wait because without physical Hardware no one can experiment solutions.

The thing here is that the Processor may want to validate the Chipset for certain features (Reason why A300/X300 were supposed to exist), and refuse to enable these features if not on supported Processor-Chipset combinations. Both Intel and AMD do that. However, on top of that, Intel uses Chipsets for market segmentation between Workstation/Server and consumer lines on the same Socket, making post-Skylake Xeons E3s to outright refuse to POST on consumer Chipsets. However, that is not a technical limitation, is a fully artificial one (And actually, it was defeated via BIOS modding, making such combinations now possible). The problem is whenever AMD decides to implement such checks in Zen 3 to not work in non-supported Chipsets instead of just crippling some problematic feature like PCIe 4 on non-X570 Motherboards.



About PCIe 4.0 on older Motherboards, that is something that was already well covered. While in theory you can, and early AGESA versions supporting Zen 2 could enable PCIe 4 capabilities (For the PCIe Slots wired to the Processor, obviously, the rest from the Chipset were still PCIe 2), PCIe 4 is quite strict and Motherboards not designed with it in mind would have signal quality issues (This seems to not have been the case with Intel Ivy Bridge, which introduced PCIe 3 support for the Processor PCIe Slots on older Sandy Bridge Motherboards). While Motherboard manufacturers could have come up with redesigned Motherboards based on old Chipsets where the Processor PCIe Slots were fully PCIe 4 compliant, AMD decided that it didn't wanted any blaming over stability issues caused by older Motherboards thus crippled PCIe 4 on all non-X570 Chipsets for that reason. I do believe that it could have been done, though, but maybe there wasn't any other way to discriminate between an old and new design based on a specific Chipset thus AMD prefered to blacklist them all to play it safe.



Also, I think that people are understimating SATA. For one, it still is much cheaper than NVMe SSDs, so you can't replace it if looking from a GB/$ perspective, and this matters for budget users (I suppose that there is a reason why there are budget Notebooks that still come with HDs!). Second, you can get far more SATA connectivity than NVMe, since NVMe is very heavy on PCIe Lane count, and to fanout PCIe, you require expensive PCIe Switches. Third, thanks to PCIe and SATA Controllers being multiplexed onto the same Pins in several chips (Including Zen itself and the AMD Chipsets), you can have fully expose both of them if a Motherboard designer decided to use an appropiated connector.
For example, X570 has 4 dedicated SATA, and 8 lanes that can be configured as either 4 SATA or 4 PCIe. If you used two OCuLink Ports, with cables you could get either a 4 lane PCIe NVMe drive or use a breakout cable for 4 SATAs. That way you could maximize flexibility and let the user could choose whatever it wants (Albeit the cables are still a non-irrelevant cost). Sadly, they all took the pathetic M.2 route, which shouldn't have scaled up from Notebooks form factors to Desktop on the first place.
 

amrnuke

Golden Member
Apr 24, 2019
1,181
1,772
136
It's official support, only "selective". As it stands now, this won't repeat with 400 series unless AMD changes their stance, which is to not provide OEMs with the necessary AGESA updates.

View attachment 20796
Can't you run the 3600 on A320?
Don't the 1600, 1800X, etc run just fine on X570?
I don't think this is anything to be taken as remotely the final word as to what's possible.
As of now we just don't have enough information to say whether you can run a, say, 4900X on B450 or X470.
 

rbk123

Senior member
Aug 22, 2006
745
348
136
Overall I think 3 years of solid upgrade paths for early adopters is still a "great bang for the buck", considering what Intel was doing for the last 10 years.
3-4 generations of support for a MB is pretty excellent, especially considering what Intel supplies. The issue with 3 or more gens really becomes tied to the other technologies in the PC - they advance as well and so even if the new CPU will work, you'll end up wanting to upgrade your MB to take advantage of the other new capabilities.
 

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
Honestly, I wonder if the motivation driving this lack of compatibility, is either:
1) OEMs just want to make more money than they have been, or
2) Some relatively unforeseen hardware incompatibility, which is likely also why B550 chipset and boards were delayed.

OEMs tried to stop Matisse from working on old chipsets completely last year (AMD stepped in to prevent that). Do you think they've had a change of heart since then?

AMD doesn't want their platform to be known as an "unstable one", they don't need that (dis-)reputation. Not now.

Matisse on old boards was a total crapshoot until about September of last year. Between that, the suicidal 3600s, and the "oh noes it only boosts to 4575 MHz" complaints, their platform already took some PR hits. Not that it mattered, since they're still selling like mad in the DiY space (at a minimum). But still.

About the compatibility, OEM wants more compatibility,

So let me get this straight: the major OEMs tried to stop Matisse from working on any chipset but x570, and yet you claim they want more compatibility? We have evidence for this courtesy of Hardware Unboxed.

Meanwhile, some A320 boards -did- get Matisse support. Eventually. My own Asus A320M-K included.

That's odd, but interesting. I'm wondering if the component-level similarities between B350 and A320 enabled them to adapt firmware. Since they got stuck supporting B350, they might as well support the other.
 
Last edited:

maddie

Diamond Member
Jul 18, 2010
4,787
4,771
136
OEMs tried to stop Matisse from working on old chipsets completely last year (AMD stepped in to prevent that). Do you think they've had a change of heart since then?



Matisse on old boards was a total crapshoot until about September of last year. Between that, the suicidal 3600s, and the "oh noes it only boosts to 4575 MHz" complaints, their platform already took some PR hits. Not that it mattered, since they're still selling like mad in the DiY space (at a minimum). But still.



So let me get this straight: the major OEMs tried to stop Matisse from working on any chipset but x570, and yet you claim they want more compatibility? We have evidence for this courtesy of Hardware Unboxed.



That's odd, but interesting. I'm wondering if the component-level similarities between B350 and A320 enabled them to adapt firmware. Since they got stuck supporting B350, they might as well support the other.
They got trapped into a situation where Intel pretty much forced motherboard changes fairly quickly with CPU generations and still commanded most of the DIY type builds. Their profit expectations and business models became designed for that, and must be under assault now that effectively motherboard/CPU sales ratios must have fallen, even if only slightly with AMD gaining marketshare.
 

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
They got trapped into a situation where Intel pretty much forced motherboard changes fairly quickly with CPU generations and still commanded most of the DIY type builds. Their profit expectations and business models became designed for that, and must be under assault now that effectively motherboard/CPU sales ratios must have fallen, even if only slightly with AMD gaining marketshare.

That seems correct. Their revenue model is based on being able to sell new motherboards frequently at high price points (MSRP has to go up as the size of the market shrinks). OEMs have likely been pushing since at least last year for AMD's ecosystem to replace the Intel revenue stream; after all, if mindfactory.de is any indicator, AMD may be selling ~90% of the CPUs in the DiY space in some markets.
 
Reactions: spursindonesia

coercitiv

Diamond Member
Jan 24, 2014
6,393
12,823
136
Can't you run the 3600 on A320?
Don't the 1600, 1800X, etc run just fine on X570?
You probably can run 3600 on A320 (ex: MSI A320 Grenade) but you cannot run Ryzen 1000 on X570. (checked with MSI Tomahawk and Asus TUF)

I don't think this is anything to be taken as remotely the final word as to what's possible.
As of now we just don't have enough information to say whether you can run a, say, 4900X on B450 or X470.
It is the final word as to what is possible within what AMD decided: Hardware Unboxed reached out to AMD and got a direct answer - they won't give OEMs the necessary AGESA update to make "selective" support possible.

So unless AMD change their mind again, the 4000 CPUs series won't run on 400 series chipsets. (as others have already said in the thread, it's probably the other way around - OEMs likely pressured AMD to draw this line)
 

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
Summit Ridge isn't supported on any x570 of which I am aware. The only reason why x570 would support 1600AF is that it is Pinnacle Ridge.
 

uzzi38

Platinum Member
Oct 16, 2019
2,702
6,405
146
Oh no, the 16th of June is not gonna be a good day for AMD unless they go back on the Zen 3 support.

Like, you thought the outrage was bad already, trust me it will be worse by that day at the latest.

I ain't sparking this storm while AMD can still go back on what they've said ahead of time.
 

CHADBOGA

Platinum Member
Mar 31, 2009
2,135
832
136
I currently have the following X370 boards. I may have some I am forgetting. MSI X370 XPower Gaming Titanium, ASUS Prime X370-Pro, ASUS ROG Crosshair VI Hero, ASRock X370 Taichi, MSI Performance Gaming B450I Gaming Plus, ASUS ROG Strix B450-I Gaming, B450 I AORUS PRO WIFI. I think they had a good run.
How come you have so many x370 boards?

What do you do with all of them?
 

randomhero

Member
Apr 28, 2020
184
251
136
I will say that this is pretty s&#@$& move by AMD. That Hallock PR piece is quite hollow. To be quite honest, I don't care what are their reason for this move.They rode that wave of customers expectations and now they are trying to weasel out.
 
Reactions: CHADBOGA

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
Meanwhile, some A320 boards -did- get Matisse support. Eventually. My own Asus A320M-K included.

https://www.asus.com/Motherboards/PRIME-A320M-K/HelpDesk_CPU/

Official 3000 series support. I don't know how they crammed everything in there but it apparently supports everything back to Bristol Ridge too.

It varies from vendor to vendor.

The most inefficient is MSI, they'd already hit a wall implementing Zen2 and had to remove Bristol Ridge support, cut down the fancy/cringy UI back to what you'd see on the first UEFI BIOSes AND remove the RAID module to make enough room. This made them release the MAX series boards with a 32MB chip to make up for their POOR software...

On the other hand, ASUS has one of the most space efficient implementations. Their 16MB BIOS images have ~12MB used out of 16MB to support everything from Bristol Ridge to Matisse.. Full user interface and everything.

This bios supports Bristol Ridge, Zen, Zen+, Zen2, Zen APUs and Zen+ APUs.

Am I missing something on how to interpret the data, or does it mean the support of all AM4 CPUs up to now, including Excavator APUs, fits in 11 MB?

If this is true, then there is no way that Zen3 would require more than 5 MB by itself... or that it would not be possible to drop Bristol Ridge support in future beta bioses in case some extra space is needed.

EDIT:

I've been checking older bioses for the X370 Prime-Pro

v1201 (Bristol Ridge, Zen): 7.0 MB

v4207 (Bristol Ridge, Zen, Zen+): 9.5 MB

v5220 (Bristol Ridge, Zen, Zen+, Zen2): 11.8 MB

I'm always using the last bios that supported up to certain generation. v1201 is the last supporting "only up to Zen", v4207 the last supporting "only up to Zen+". I do not know exactly which would be the limits for the APUs.

It seems that every generation requires adding ~2.3-2.4 MB more than the previous. If the Zen3 microcode is approximately the same weight, you could still fit 2 generations in this bios chip.

Also giving that ASROCK owners claim their Zen2 bioses weight ~8 MB, I'm inclined to say that ASUS bios UI is probably sort-of bloated and could be stripped down a bit if needed.

Asrock seems to have an even more efficient implementation going by that post.

The 16MB BIOS chip size excuse is BS, anyone could see through the lies. Of all possible valid technical excuses, they chose the least technical and most stupid one that also incriminates board partners.

Gigabyte's entire X570 lineup comes with a 16MB chip, lol.

Make no mistake, this is still light years away better from what we're used to from Intel, but still. Way to shoot yourself in the foot, and undermine most of the goodwill you've managed to build up from the brink of bankruptcy in a single moment.

-------------------------------

BTW, Zen1 support on X570 through AGESA 1.0.0.4b is as official as it can get. AMD themselves made it so by publicly talking about it. It was unofficial on AGESA 1.0.0.3x variants, you could run 1st gen processors on these versions + X570 when the official slide clearly stated no support. (at least on Asrock, ASUS boards).


On B550, I just hope that when these are released, they're cheap enough to be an insignificant cost if one wants to go Zen3 and keep their DDR4. If it's enough of an upgrade over Zen2 to even make these users want to upgrade, they can always "fork" their current rig by getting some more DDR4 and a couple more parts. That's neat.
 
Last edited:

Insert_Nickname

Diamond Member
May 6, 2012
4,971
1,692
136
On the other hand, ASUS has one of the most space efficient implementations. Their 16MB BIOS images have ~12MB used out of 16MB to support everything from Bristol Ridge to Matisse.. Full user interface and everything.

Thank for the link. 12MB for full AM4 + GUI is impressive.

I suppose you could fork the BIOS if required, so you could perhaps fit Renoir and Vermeer support by taking away Bristol Ridge and Summit Ridge. Then make a "legacy compatibility" BIOS version for system that need it.

BTW, Zen1 support on X570 through AGESA 1.0.0.4b is as official as it can get. AMD themselves made it so by publicly talking about it. It was unofficial on AGESA 1.0.0.3x variants, you could run 1st gen processors on these versions + X570 when the official slide clearly stated no support. (at least on Asrock, ASUS boards).

Has anyone tried yet? Are there reports of it working in the wild?
 

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
Thank for the link. 12MB for full AM4 + GUI is impressive.

I suppose you could fork the BIOS if required, so you could perhaps fit Renoir and Vermeer support by taking away Bristol Ridge and Summit Ridge. Then make a "legacy compatibility" BIOS version for system that need it.

This is another nail in the coffin for the 16MB chip too small excuse. Boards that have flashback don't even have an excuse for not having this option available for them, you can't brick the board nor be left with an unbootable system if you flashed whatever without checking what CPU you started with first.

Has anyone tried yet? Are there reports of it working in the wild?

Have a read.

From that post, the AORUS X570 Master can boot the 1700 on BIOS F4... which is based on AGESA 1.0.0.3. MSI's X570 A Pro can, too, on the first BIOS.

No Zen1 on X570 at launch according to the official slide. Sure...

------------

I can personally back that up.

A friend had a non AF 1600 + B350 board, and got an asrock x570 board (Phantom Gaming 4) for cheap, he couldn't pass on that deal. He didn't have any other CPU at that moment. Told him he got lucky as 1.0.0.4b had just been released for that board including official Zen1 support, as it has flashback he could update it and get going, but he tried it out of the box.

He sends me this:



800F11/8001138 is Summit Ridge's (Zen1) latest microcode. I was like, wut

I then go on my phone to Asrock's page for that board...



As you can see, this BIOS definitely includes Zen1 microcode:



So... Yeah. It booted on <1.0.0.3x AGESA. On X570. He told me it worked mostly fine, some weirdness on the USB ports that come from the CPU misbehaving sometimes, but otherwise it was usable. The 1.0.0.4b BIOS solved these issues. He still sold it at some point and got a 2600 for cheap later.

Don't believe what AMD's official slides tell you about official CPU and chipset compatibility. They weren't right about Zen2 and 300 series boards (we now have A320 boards running the 3950x), they weren't right on X570 and Zen1 (at least for Asrock).

These CPUs are SoCs (remember A300 that nobody bothered with? It was just the AM4 socket), the motherboard merely provides connectivity.

------------------

IMO, There is no excuse for this save for business reasons. All this Zen3 thing will end up backfiring for whoever decided to take this last minute decision to screw X470 and below users. Either AMD or some adventurous OEM or the community will get Zen3 AGESA working on these boards, at least the popular ones.
 
Last edited:

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
@.vodka

It would be better if AMD just came out and told the truth instead of talking about ROM sizes, though in MSI's case, they may actually be right. There's so much inconsistency between Asus and MSI here that it's not hard to understand why AMD doesn't want to mess with this headache anymore. That and the level of proper Matisse support on 300 and 400-series boards last year was pretty bad @ launch. It took a few months to hammer out that situation. The lack of transparency here from AMD is disheartening. Kind of makes you wonder why they feel the need to pave over the issue. Whom exactly are they protecting?
 

uzzi38

Platinum Member
Oct 16, 2019
2,702
6,405
146

IMO Hardware Unboxed nailed the problem with the situation and why it's even come about. It's not an issue with AMD being greedy which shouldn't come as a surprise as they don't really make anything from each chipset sale. It was always just about AMD not willing to put in the time and effort to support board partners through the process and not wanting to deal with the poor reaction from the community when thing inevitably don't work straight away.
 

piokos

Senior member
Nov 2, 2018
554
206
86
It would be better if AMD just came out and told the truth instead of talking about ROM sizes, though in MSI's case, they may actually be right.
What would be the actual workflow to check CPU-chipset compatibility on AMD website? Where is it?

Of course everyone should follow mobo maker recommendation, but since AMD decided to mention this on slides...
Don't believe what AMD's official slides tell you about official CPU and chipset compatibility. They weren't right about Zen2 and 300 series boards (we now have A320 boards running the 3950x), they weren't right on X570 and Zen1 (at least for Asrock).
If slides are incorrect, where is the proper source?
Because the whole point of "official compatibility" is that it's what the manufacturer says - not what actually works (even if always).

So if 3950X seems to work on a A320, that would *at best* be unofficial compatibility.
But it could be partial compatibility (not everything works) or just a coincidence (it works on some boards etc).

The whole point of "official compatibility" is not about learning what *may* work (you can learn that from the community), but what is *guaranteed* to work.

With a bit of BIOS modification you can run Coffee Lake on Z170. It doesn't mean you should.
 

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
What would be the actual workflow to check CPU-chipset compatibility on AMD website? Where is it?

At this point, I don't think you can fully trust AMD's slides or website. Obviously some boards defy expectations. You'll have to dig deep and talk to people who own/use the boards in question.
 

SPBHM

Diamond Member
Sep 12, 2012
5,058
410
126
very poor decision by AMD, it's the same socket, electrically the CPUs will be compatible, just lacking the firmware support, which for a company that is designing CPUs and GPUs to the level of complexity they are can't be an impossible task, the bios size excuse is clearly not good enough.

as well mentioned b450 is still being sold new everywhere and is the popular choice for Zen2, due to AMD taking forever to release b550 in great part, which makes it even worse.

and x570 looks like a flawed solution with the high power usage/active cooling requirements and decreased sata performance and high cost..

at the end of the day it's better than what Intel does I guess, but they could've done/planned things better, and been more transparent.
 

piokos

Senior member
Nov 2, 2018
554
206
86
At this point, I don't think you can fully trust AMD's slides or website. Obviously some boards defy expectations. You'll have to dig deep and talk to people who own/use the boards in question.
As long as those people will replace my broken hardware...

But seriously, not even all DIY customers will accept having to ask around on forums.

I wonder if AMD provides proper documentation to AIBs/OEMs or if they have to reverse-engineer and test everything on their own.
Either way, currently they're probably taking this risk on themselves. Commendable but sad.
 

Insert_Nickname

Diamond Member
May 6, 2012
4,971
1,692
136
So if 3950X seems to work on a A320, that would *at best* be unofficial compatibility.
But it could be partial compatibility (not everything works) or just a coincidence (it works on some boards etc).

The whole point of "official compatibility" is not about learning what *may* work (you can learn that from the community), but what is *guaranteed* to work.

It may be as simple as some Matisse/Vermeer features gets disabled or downgraded. F.x. Matisse has 10Gbit USB support directly from the CPU, but not when using a non-X570/B550/A520 board.

It's just a guess, but AMD may be referring to full feature support. If you want ALL platform/CPU features, Matisse is only compatible with X570/B550/A520. Then they're correct. If you don't mind, sure you can run it on any AM4 chipset.
 

amrnuke

Golden Member
Apr 24, 2019
1,181
1,772
136
As long as those people will replace my broken hardware...
That's the only level of disappointment here for me if Cezanne does work on 4x0 boards but only unofficially. If Zen3 works on B450, great, but if it does so in an officially unsupported manner, then that has a lot of drawbacks for people who use such products for professional environments (like on my office's workstations that require a little more grunt).

I am actually fairly optimistic motherboard makers will bake in "alpha" support for Zen3 on 4x0 chipsets (with AMD's blessing) but strictly prohibit warranty replacement or support should something go wrong.
 

piokos

Senior member
Nov 2, 2018
554
206
86
It's just a guess, but AMD may be referring to full feature support. If you want ALL platform/CPU features, Matisse is only compatible with X570/B550/A520. Then they're correct. If you don't mind, sure you can run it on any AM4 chipset.
So they should explicitly say what works and what doesn't. Just like they openly say how PCIe4.0 support looks depending on platform.

How am I supposed to know this otherwise?
 
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/    |