Opera announces 'gradual transition' to WebKit for desktop and mobi


Recommended Posts

Personally I liked Opera developers efforts for Presto engine but they had little user base so cost benefit was less. Webkit is free open source contributed by many many people so it will good for them switch but it will be monopoly. Which I hate...

Personally I liked Opera developers efforts for Presto engine but they had little user base so cost benefit was less. Webkit is free open source contributed by many many people so it will good for them switch but it will be monopoly. Which I hate...

Monopoly is really in regards to one company. If Chromium for example becomes the standard it won't be a monopoly as many companies and people contribute to it.

1st Step: Get everyone on Chromium/Webkit

2nd Step: Get it to be official standard where everyone contributes to and is tightly connected with W3C

3rd Step: Enjoy rapid innovation and progress on the web without waiting a decade for new features to trickle down and deal with different browsers that wanted to integrate and interpret the HTML/CSS/JS standards however they want.

There's a reason why we are still waiting on ECMAScript 6

Look at this mess:

http://kangax.github.com/es5-compat-table/es6/

Monopoly is really in regards to one company. If Chromium for example becomes the standard it won't be a monopoly as many companies and people contribute to it.

1st Step: Get everyone on Chromium/Webkit

2nd Step: Get it to be official standard where everyone contributes to and is tightly connected with W3C

3rd Step: Enjoy rapid innovation and progress on the web without waiting a decade for new features to trickle down and deal with different browsers that wanted to integrate and interpret the HTML/CSS/JS standards however they want.

There's a reason why we are still waiting on ECMAScript 6

Look at this mess:

http://kangax.github...mpat-table/es6/

No actually as some one on previous pages mentioned that sometime user contribution is ignored because other does not want in case of Webkit then this is sad fact. I am personally not a follower of Webkit or Chrome development except I read Chromium changelogs while I do follow Mozilla development quite much. Again Microsoft is quite silent and secretive except their Microsoft Connect for feedback.

Lets hope whatever happen, it will be in favour of developers, users both at a time.

BTW one general question? Whether they will go with Webkit or Webkit2 API, Webkit 2 API is same as Chrome multiprocess architecture (as far as I know) and used in Safari 5.2+ which never released for Windows? Am I right? So which one Opera choice?

I suggest Webkit 2 API..

Monopoly is really in regards to one company. If Chromium for example becomes the standard it won't be a monopoly as many companies and people contribute to it.

1st Step: Get everyone on Chromium/Webkit

2nd Step: Get it to be official standard where everyone contributes to and is tightly connected with W3C

3rd Step: Enjoy rapid innovation and progress on the web without waiting a decade for new features to trickle down and deal with different browsers that wanted to integrate and interpret the HTML/CSS/JS standards however they want.

There's a reason why we are still waiting on ECMAScript 6

Look at this mess:

http://kangax.github.com/es5-compat-table/es6/

I see you still don't understand the difference between a standard and software. It's important to have a standard and it let software be the standard. It stifles innovation, even when OS and breaks thievery purpose of a standard.

BTW one general question? Whether they will go with Webkit or Webkit2 API, Webkit 2 API is same as Chrome multiprocess architecture (as far as I know) and used in Safari 5.2+ which never released for Windows? Am I right? So which one Opera choice?

I suggest Webkit 2 API..

from the wording in the article, it sounds like opera is going to be based on chromium.

Well exactly.. that's the whole point.. if we had that on desktop it would 10x better than what we have now.

No it wouldn't... and I like how you ignored everything else in my post, par for the course of course when it comes to you.

No it wouldn't... and I like how you ignored everything else in my post, par for the course of course when it comes to you.

Developing HTML5 stuff for mobile where we really have two major webkit based browsers is 10 times easier than for desktop. I read your whole post, this FACT disproves your opinion.

Developing HTML5 stuff for mobile where we really have two major webkit based browsers is 10 times easier than for desktop. I read your whole post, this FACT disproves your opinion.

The easiest is of course to develop using the STANDARD, not using the prefixes at all until the prefixed function are actually standardized.

the only thing you're accomplishing here is giving away control of the web, AGAIN, and hindering innovation, AGAIN.

The fact you can't code and write the same code 15 times for no good reason is besides the fact. and the reason it's easier to code for mobile isn't because 90% of mobile internet users are on iOS/Webkit/Safar. it's because they'r eon mobile and mobile sites by default are simpler and easier.

omg it's easier to put together a 20 piece puzzle than a 500 pieze one, THE REVELATION!!!

It also doesn't help that the majority of code added to websites for browser specific functions are not there to ad functionality, but to falsely detect the browser(which is the wrong way anyway, you're supposed to detect what the browser supports, not the browser, but as a coder I suppose you know this...), and then most of the code for this again, is simply to push broken code to certain browsers. I mean look at the amount of webpages that suddenly work perfectly even better than in Chrome if you let opera pretend to be chrome.

The great holy google did this themselves right after they bought google docs. their very first update(this was before Chrome when IE had 90% market and FF was heavily paid by Google), the very first google update to the "google docs" before they even changed the name only did one thing. It used deep browser scanning to detect Opera(so changing the user agent wouldn't work) and sent broken code to it. there was NO other changes. just breaking the webapps in Opera, and suggestign they switch to the google supported FF.

from the wording in the article, it sounds like opera is going to be based on chromium.

So Webkit1 with Chrome like multiprocess architecture. Thanks for pointing my thinking at that point as well.

BTW one general question? Whether they will go with Webkit or Webkit2 API, Webkit 2 API is same as Chrome multiprocess architecture (as far as I know) and used in Safari 5.2+ which never released for Windows? Am I right? So which one Opera choice?

I suggest Webkit 2 API..

WebKit2 and Chromium (Chrome) aren't the same. Chromium is based on WebKit "1" and Google developed their own multiprocess & sandbox environment and slapped it on top of it. WebKit2 is built differently, but offer similar functionality.

In my experience, while testing Safari and Chrome, is that WebKit2 seems to manage processes better (memory) but apart from that I couldn't really say... I'm by no means an expert in the area and I couldn't talk about/compare their security to save my soul.

I wish Opera had chosen WebKit2 in any case, but it seems to me quite clear that they're going to fork Chromium. It'll be interesting either way though. Opera have innovated so much already!

it's easier to code for mobile isn't because 90% of mobile internet users are on iOS/Webkit/Safar. it's because they'r eon mobile and mobile sites by default are simpler and easier.

omg it's easier to put together a 20 piece puzzle than a 500 pieze one, THE REVELATION!!!

Absolutely wrong..

Have you written serious web apps for mobile? Well I have.

The reason why it's easier to write is because of Webkit.. we have most of the features unified (with some slight differences that Apple added on iOS Safari but are really not deal breaking - aka you don't have to use them and have mostly to do with scaling and zooming in/out pages). The only real big difference is that Apple uses Nitro JS VM while Google uses V8 so there might be some discrepancy in JS performance but that's not really much of an issue as well because they are both pretty fast.

The reason HTML5 actually started getting used more and more is exactly BECAUSE of mobile. The first and more advanced web apps have been written for mobile first because the HTML5 features have been added rapidly for both Android and iOS. Both Android and iOS browsers have been able to innovate very quickly and bring new features and fix issues that didn't work.

CSS3 stuff has been working on both since pretty much the beginning when no desktop browser really supported it (I built things in 2008 for iOS that used cool 3D effects and crap). The only difference between the two were that Google got a bit later in to have hardware acceleration on CSS3 transitions and stuff but that was because of the OS and not really related with webkit.

Today even, HTML5 is hardly used on the desktop because of the mess it is in, however people have started building full blown apps with HTML5 on mobile and they are now combining with tools that package that code along with hardware APIs so we can access specific device capabilities (aka PhoneGap and others).

You won't see this ubiquity on the desktop web for a while. If we are lucky in 5 years. Which would make it about 10-11 years since HTML5 officially got to W3C/WHATWG.

Hell, even WebGL is now being introduced in Chrome on Android (via --enable-webgl flag - experimental) and most likely in Safari soon as well when Google officially pushes it.

And it's going to have 90%+ support on mobile years before it will have that ubiquity on desktop (if ever).

I'm not sure what you are talking about when you say it's not easier.. it's 10 times easier. It's nicer for development too. Less waste of time and unnecessary code just to make your web app appear and work the same across browser. It's really a joy to develop for unlike for desktop.

Today even, HTML5 is hardly used on the desktop because of the mess it is in, however people have started building full blown apps with HTML5 on mobile and they are now combining with tools that package that code along with hardware APIs so we can access specific device capabilities (aka PhoneGap and others).

:huh:

PhoneGap (owned by Adobe), other tools owned by other companies, and you think this is a good thing?

HTML is a markup language, and you can barely find a good agreement on how to just write a document that gets presented with standard markup that has a logical semantic outline, yet some companies want it to flash and dance and work as some magical do-it-all client side language which is a terrible idea. Five years out we're going to have HTML the document language, and HTML/js/CSS the application development language, which is fragmented but works on every device if you don't mind kludging it together with 3rd party tools.

WebKit2 and Chromium (Chrome) aren't the same. Chromium is based on WebKit "1" and Google developed their own multiprocess & sandbox environment and slapped it on top of it. WebKit2 is built differently, but offer similar functionality.

In my experience, while testing Safari and Chrome, is that WebKit2 seems to manage processes better (memory) but apart from that I couldn't really say... I'm by no means an expert in the area and I couldn't talk about/compare their security to save my soul.

I wish Opera had chosen WebKit2 in any case, but it seems to me quite clear that they're going to fork Chromium. It'll be interesting either way though. Opera have innovated so much already!

Thanks for information but I already know this though. Also same wish that Opera should go with Webkit 2 but as pointed by ViperAFK, they might go with Chrome approach.

So Webkit1 with Chrome like multiprocess architecture. Thanks for pointing my thinking at that point as well.

Multiprocess can be done with any rendering engine and has nothing to do with the actual rendering engine. It's the actual program and how that is programmed that defines that. hopefully Opera decides to do it as they currently do, multi process, but wrapped inside the opera file so your task manager isn't flooded by 50 opera.exe files. god that's annoying in chrome.

Personally I much prefer a well programmed browser like opera that doesn't crash, as opposed to a browser like chrome where the crashes are separated. to not affect the whole browser.

I'm not sure what you are talking about when you say it's not easier.. it's 10 times easier. It's nicer for development too. Less waste of time and unnecessary code just to make your web app appear and work the same across browser. It's really a joy to develop for unlike for desktop.

It's also 10 times easier to JUST STICK TO THE FRICKEN STANDARD, and don't use the damn prefixes. less resources, better code, faster site, cleaner code. and far less effort.

Just because you're incapable of coding without adding in 50 fancy and completely unnecessary "cool" effects that are only supported on some browsers, doesn't make it harder, it just means you're unable to to not add in lots of unnecessary CSS code that makes your life harder.

Also you're not supposed to code for browser, you're supposed to use the standard function for checking what is supported in the browser visiting the site an activate CSS effects based on that. easy, and clean on the code and efficient. as opposed to checking for the browser and manually adding code for each browser and each version of the browser.

Also as for VRML2... err I mean, WebGL.... bwahaha. sorry that crap isn't and shouldn't be used, EVER. it makes horribly designed 100% flash sites look like user friendly interface masterpieces. and it's horrible to use for games as well. better to use an actual plugin for that, r an actual app. but hey you're completely against separate apps like flash and think html5 is the greatest best thing ever... err wait a min now...

  • Like 3

Monopoly is really in regards to one company. If Chromium for example becomes the standard it won't be a monopoly as many companies and people contribute to it.

1st Step: Get everyone on Chromium/Webkit

2nd Step: Get it to be official standard where everyone contributes to and is tightly connected with W3C

3rd Step: Enjoy rapid innovation and progress on the web without waiting a decade for new features to trickle down and deal with different browsers that wanted to integrate and interpret the HTML/CSS/JS standards however they want.

Funny, I have a deja-vu feeling. It was 20 years ago with UNIX and all the standards like POSIX or CDE, when Sun, IBM, Digital, HP and SGI were all battling while promising a better standardized UNIX future.

I heard the same kind of arguments from developers who were working in the comfort zone of SunOS and did not want anything to do with Irix because it was different

We all know how it ended.

Multiprocess can be done with any rendering engine and has nothing to do with the actual rendering engine. It's the actual program and how that is programmed that defines that. hopefully Opera decides to do it as they currently do, multi process, but wrapped inside the opera file so your task manager isn't flooded by 50 opera.exe files. god that's annoying in chrome.

Personally I much prefer a well programmed browser like opera that doesn't crash, as opposed to a browser like chrome where the crashes are separated. to not affect the whole browser.

If it is not at engine level in case of Webkit but I think Gecko is different. Where Mozilla ditched idea of E10S (electrolysis|) project and now they are doing snappy project by moving code to aysnc and from away from main thread.

Lets hope whatever we get makes web developers and users both happy.

This is awful news.

Well exactly.. that's the whole point.. if we had that on desktop it would 10x better than what we have now.

Just like 90% of people using IE6, right? Jesus Christ your lack of logic is hurting my brain.

  • Like 2

Personally I much prefer a well programmed browser like opera that doesn't crash, as opposed to a browser like chrome where the crashes are separated. to not affect the whole browser.

Spot on. ;)

Know whats funny? Boz still hasn't answered to Athenar's challenge but keeps repeating the same crap he always does.

He hasn't answered anyone challenges and merely selectively quotes 1/10th of people's posts when he replies or just ignores them altogether.

So you're saying Webkit is a closed source project now? Or maybe that "It's fine because it's open source" defence doesn't work out so well in practice?

He's saying that Google doesn't own Webkit. Google can't just do whatever they want to with it.

This is awful news.

Why?

He's saying that Google doesn't own Webkit. Google can't just do whatever they want to with it.

They can, and then we have WebKit fragmentation and... Square one

Again, software is not a standard. Standard is not software. Software follows standards. Standards are written for software.

  • Like 2
This topic is now closed to further replies.
  • Posts

    • These are not "great" prices... just "less awful". Apparently "Those who forget the past are doomed to pay higher prices and think they've won."
    • Russia was able to invade Crimea because of those people. But my point is that I've personally heard how great it was to be "back in Russia" right afterwards - look how great it is now. I've asked you a question in another comment which you haven't answered, so I'll ask it again: is it better now without "Europrats"?
    • ZimaBoard 2 1664 Starter Kit review: it's a cool and affordable DIY NAS by Steven Parker IceWhale Technology reached out to me asking if I was interested in testing the ZimaBoard 2, and after convincing them to send me the Starter Kit, it arrived at my doorstep in May. A bit of background: it is a Shanghai-based Chinese company founded in 2020, which specializes in single-board servers and personal cloud solutions. From searching around online, user feedback on the company and ZimaOS is mostly positive, so we're off to a good start. In addition, I should probably point out that although they do not have a large portfolio of NAS devices, with just four of what they do offer, they seem to have covered everything from a relatively low-priced entry point with the ZimaBoard 2, right up to the high end, with the ZimaCube 2 Creator Pack that even includes an NVIDIA RTX PRO 2000. Anyway, as already mentioned, what we have today is the ZimaBoard 2 Starter Kit, and here are the full specifications: ZimaBoard 2 Model 832, 1664 CPU Intel Core N150 (4x E Cores/Threads, Max burst up to 3.6 GHz) TDP: 6W (Base) 10W (Max) Graphics Intel UHD Graphics 24 EUs (1.00 GHz) Memory 8 GB, 16 GB DDR5 4800MT/s non ECC SODIMM (soldered) Disk Capacity 60 TB (30 TB x 2) Supported RAID Types TRAID, TRAID +, RAID0, RAID1, RAID5, RAID 6, RAID 10 Storage 2 x SATA 3.0 6Gb/s Ports with Power Bootloader 32 GB, 64 GB eMMC Network 2x RJ-45 2.5 GbE PCIe 1 x PCIe 3.0 (via LPC) USB Ports 2 x USB-A 3.1 (5 Gbps) Display Mini-DisplayPort 1.4 (4K@60Hz) Hardware Transcoding Engine H.264, H.265, MPEG-4, VC-1 Maximum resolution: 4K (4096 x 2160); Maximum FPS: 60 Virtualization Intel® AES New Instructions Intel® Virtualization Technology (VT-x) Intel® Virtualization Technology for Directed I/O (VT-d) Size (H/W/D) 140mm x 83mm x 31mm Weight 0.4 kg (only ZimaBoard 2 device) Power 12v 5A Power Supply Warranty 1 year (Global) 2 Years (EU) OS ZimaOS v1.6.1 MSRP $339, $399 ($548.60) As you can see above, there are two variants of the ZimaBoard 2. The lesser variant has half the eMMC storage and 8 GB less RAM, although it also costs $60 less than the top variant we are testing today. The above pricing is only for the ZimaBoard 2. I put the MSRP of the Starter Kit next to it in brackets, although as of publishing, it is discounted to $534.50. The ZimaBoard 2 started life on Kickstarter and shipped to backers in August last year. It became available via the official website in late 2025 and Q1 2026. This hobbyist NAS contains the still relatively new N150 Intel CPU released in the first quarter of 2025, with support for DisplayPort 1.4, HDMI 2.1, although in this case, the memory is integrated into the board itself, so it will not be possible to upgrade or expand the amount. It also supports AV1 decoding, as well as H.264, VP8, VP9, H.265 (8 bit), and H.265 (10 bit). The different capabilities in the Alder Lake-N (and Twin Lake) series are listed below. Processor E-cores L3-cache Turbo clock GPU GPU-clock TDP Intel N355 8 6 MB 3.9 GHz 32 EUs 1.35 GHz 9 W Intel Core 3 N350 3.9 GHz 1.35 GHz 7 W Intel Core i3-N305 3.8 GHz 1.25 GHz 9 W Intel Core i3-N300 3.8 GHz 1.25 GHz Intel N250 4 3.8 GHz 1.25 GHz 6 W Intel Processor N200 3.7 GHz 0.75 GHz Intel N150 3.6 GHz 24 EUs 1 GHz Intel N97 1.2 GHz 12 W Intel Processor N100 3.4 GHz 0.75 GHz 6 W The CPU is part of the Twin Lake series that sits near the bottom of the N-series, designed for low- powered systems and entry-level laptops, and as such has a base level TDP of just 6W. As I have noted before, we are seeing another NAS with a great amount of RAM. It's important to mention that the ZimaBoard 2's memory is integrated into the base board (which is why they have two variants of it). As a reminder, up until a couple of years ago, it was commonplace to only get 2 or 4GB max on a flagship Synology or QNAP home NAS. Ever since the likes of TerraMaster and more have entered the market with ample RAM sizes included in their NAS offerings, it has gone a long way in forcing the hands of the traditional makers to up their game a bit. First impressions The Starter Kit came in one outer box with several packages inside it (shown above). I forgot to take pics of it because when it arrived, it wasn't clear what was inside, and I had to confirm with my contact that I received the entire Starter Kit. In the box ZimaBoard 2 ZimaBoard 2 HDD Expansion Bracket + PCIe card frame Zimaboard Mini DisplayPort Male to HDMI Female Cable 4K 60Hz Zimaboard PCIe 3.0 x4 to Dual NVMe M.2 SSD Adapter Card Quick guide [full online guide] Limited warranty notice Screws Design Where to start? You'd be forgiven for mistaking it as an SSD enclosure if not for all the ports on it. It is completely made out of metal, and the top is an entire heatsink. It has a premium feel about it, but it definitely looks like a hobby device. As you will see, the completed build looks like it belongs in a server or meter closet rather than as a showpiece on someone's desk. On what I am calling the rear, there's a Mini DisplayPort (1.4), two 2.5 GbE ports, with Type A 3.1 USB ports, and then the barrel connector port. Around the front, there are two SATA6 ports with a power connector in the middle. Left side Right side One side is completely free of ports. On the other there's a slit that allows for the feed of a CPU fan cable, and a PCIe 3.0 X4 slot. Top Bottom The top is entirely made up of a heatsink except for the extended height for the I/O on the rear. Around the other side, you can find the ZIMA branding and some regulatory information stamped near the bottom. As you may see from the bottom of the ZimaBoard 2, it scratches quite easily from just moving it around on my Ikea island. Teardown Before we get started, let's have a look at this thing on the inside. The steps to get to the board are as follows: Remove the four smaller Torx screws on the bottom of the ZimaBoard 2; Remove the four larger Torx screws on the sides of the device; Carefully unstick the CMOS battery from the PCB; Remove two Phillips screws on the PCB; Lift out the PCB. Yes, as you can tell from the instructions, you need three different tools to remove Torx and Phillips screws (10 in total), and unhelpfully, one of the screws is located under the CMOS battery, which is stuck onto the PCB. Building Now comes the fun part. Because the ZimaSpace website does not provide any guidance on how to put the Starter Kit together. They only have guidance for connecting the CPU fan. However, they did upload a video to their YouTube channel that shows the entire process. To install the fan, first remove the four screws on the bottom of the ZimaBoard 2, then on the inside, there is a CPU FAN connector where you can attach the fan, reattach the ZimaBoard 2 frame, and feed the fan cable through the provided slit. Then remove the nearest screw on the side and attach the fan frame to the side of the device using the same screw. ZimaBard 2 screws Aligning the screws Bottom view Remember those four screws we removed to access the CPU FAN? Longer screws are provided in the box with the HDD Expansion Bracket, which is what you will now need to attach the ZimaBoard 2 to it. Helpfully, the orientation on how to attach it is made obvious when the frame can only be screwed on at the same overall length as the ZimaBoard 2. If you do it the wrong way around (which is what I did initially) one side hangs off the frame, and it becomes difficult to attach the PCIe Adapter Card cable. PCIe card frame Other side PCIe slot connector Next, it's time to attach the PCIe card frame, which is fastened with the help of 3.5-inch SATA HDD (3 screws). These are toolless screws that you can just use your fingers to fasten them with. Then it is time to connect the provided PCIe cable with the slot connector on one side of the ZimaBoard 2, feed it through the bottom of the HDD frame, and fasten it with two standoffs. Both bracket options 2280 standoffs with 2x 4TB MP44Q The PCIe 3.0 X4 card comes with a short bracket option, handy if you decide to place it inside a different NAS or rack server, but here we need the long bracket. Oddly enough, the M.2 standoffs were preinstalled into the 22110 position, but extra standoffs are included in the box, which I installed at the 2280 position for our use. I added a couple of MP44Q M.2 PCIe 4.0 SSDs (2 x 4TB) that can be availed on Amazon for $478.99 (the lowest price for 3 months) that TEAMGROUP supplied us with Then we have the almost completed build, you just need to push the card into the PCIe slot. Unfortunately, IceWhale Technologies did not provide a screw for the PCIe card frame (this is also apparent in their own video). Here it is at several different angles, with the last pic showing the SATA Y-Cable connected to the two WD Red Plus 4TB drives. Setup and Usage Next, you connect your cables to the I/O, and the ZimaBoard 2 powers on automatically, as there is no power button on the device. Power is controlled through the Settings in ZimaOS. BIOS The ZimaBoard 2 includes an Aptio BIOS from American Megatrends [1, 2, 3], and you can setup pretty much everything here including the boot order, which is locked to the UEFI OS, however above that choice you can enable or disable booting to a SATA/USB bootloader so this would still allow you to switch to an alternative bootloader and boot from it, or disable it to instead always start from the first disk with an OS installed on it. Initial Setup Upon connecting to the LAN and booting up, the ZimaBoard 2 can be reached by navigating to the IP address (shown if you have a monitor connected), or you can find it using the ZIMA Client desktop application, which is essentially a Zima device finder. Initializing the ZimaBoard 2 The ZimaOS setup process is pretty straightforward, through a wizard, and in full above, it basically consists of setting up an account and some handy tips, and that's that! Post Setup (ZimaOS update) Upon first boot, you are alerted that there is a ZimaOS update from 1.5.0 to 1.6.1, which I applied; the full process is shown above with the changelog. ZimaBoard 2 Storage Setup Next, it is time to set up the storage. ZimaOS actually throws everything onto the eMMC flash drive; it is also the default location of AppData, which is definitely something to be wary about, as the 45GB available storage could fill up quickly. HDDs I first attempted to create a Storage Pool using the two 4TB WD Red Plus NAS drives, and got an error message: After several attempts and then looking online, I discovered it was a bug with ZimaOS where the fix was simply to reboot ZimaOS and then try again, this time I was able to create a RAID mirror using the two drives. SSDs I did the same for the SSDs, as you will see in the above gallery, when I created the second Storage Pool, it only allowed me to select available drives. ZimaBoard 2 AppData ZimaOS comes with an App Store that includes a repository of almost 400 apps, so you will be able to find most of what you'll need for a NAS (although after a quick search, I wasn't able to find a Surveillance Manager), and now comes the important part: moving the default AppData location off the 45GB eMMC and onto a larger volume: Open Settings Then Apps Then, in the Select a new location field, click on the new Storage volume you want to move it to (in my case, the Apps Storage Pool), which is the SSD RAID mirror. Confirm the Migration warning Be praised! You can also do this for Docker (which by default installs onto the 45GB eMMC flash drive) and the User database. Plex Setup Next, I tested the configuration by installing the Plex Server app from the App Store. The library folders must already exist (which I placed into the Storage Pool). Plex Server setup is straightforward and requires very little configuration. In my case, all I had to do was add the media path I just created, which you can also browse to using the folder icon in the path field. In addition, you can now map the new Media library in Windows Explorer using the Zima Client. Oddly enough, it is not possible to access the ZimaBoard 2 over the Network Neighborhood; you must map drives using the client, which is shown in the last image in the above gallery. I watched one of my Blu-Ray rips, which is Dolby Vision with Dolby Atmos, and the content played fine with no stuttering or buffering, which is what anyone needs in this scenario. ZimaBoard 2 Zima Client mobile app There's also a client for mobile. It is pretty barebones, as shown in the above gallery, for example, the Apps screen launches the WebUI for that app, and the Backup must be done manually. On opening Backup, you can select internal storage folders on your phone to backup to the ZimaBoard 2's storage, and although this is constantly scanned, the backup action itself must be manually triggered. There is an option to allow foreground backup (last image in the above gallery), but this basically means the queued backup gets triggered when you manually open the app. Benchmarking SATA PCIe 3.0 X4 A CrystalDiskMark test on a mapped network drive from within a Windows 11 25H2 PC (image above) connected over a 2.5 GbE was well within acceptable ranges. Writes were generally better on the SSD RAID mirror. SATA PCIe 3.0 X1 I also ran the NAS Performance tester, which tests the link speed performance. As you can see, it pretty much maxes out the 2.5GbE connection. Of course, you can also opt to bond the two 2.5 GbE connections for a bit more umph, but I didn't do that. Thermals Top PCIe card SATA HDDs Next, I measured some hotspots while playing content on Plex. It's fair to say this will perform better than a NAS that is enclosed in a metal or plastic case, as almost everything storage-wise is exposed! Anyway, the ZimaBoard 2 did not break a sweat with Plex streaming or disk benchmarks. ZimaOS Factory Reset ZimaOS does not include a factory reset option. Instead, you have to download the ZimaOS image and flash it to the eMMC manually. The flashing process is shown in the above gallery. The steps to do so are listed below: Download the ZimaOS image here; Open BalenaEtcher (Run as Administrator) and select the image; Select your inserted USB drive (min 8 GB) Flash to it; Connect your USB drive, monitor, keyboard, USB hub (optional), mouse (optional), and network cable (recommended) to the ZimaBoard 2; Connect power and press F11 continuously; Select your USB drive starting with UEFI in the boot device menu; Press Enter on the Install ZimaOS option; Select /dev/mmcblk0 (MMC) flash drive as target; Confirm with (three times) to wipe the target disk; Wait a couple of minutes while ZimaOS installs; Remove the USB drive and confirm with a reboot; Your ZimaBoard 2 has been factory reset. However, you don't have to stick with ZimaOS, in fact the company also offers official CasaOS images, that are based on Debian; or as they say themselves, put anything you want on this "hackable single board server" it's up to you. Conclusion I had a lot of fun putting this together. I've custom-built all my own PCs and servers since the 90s, and this is the first time I have had to put a NAS together. Even if the actual base ZimaBoard 2 was already a completed build, it still feels pretty custom. I just wish that IceWhale Technology included a getting-started guide in the box for the Start Kit, which would have really completed this kit. Instead, I had to search for the official video on the YouTube channel to make sure I wasn't doing anything wrong. So who is this for? Definitely the hobbyist who is comfortable building their own PC and servers. It also has a much smaller footprint than its nearest equivalent (in terms of specs), like the Beelink Me Pro, which is another NAS I will be testing soon. Although the Beelink does not come with the PCIe 3.0 X4 expansion, the ZimaBoard 2 Starter Kit suddenly looks to be a great bargain, even if it only offers the two 3.5-inch bays over the four in the other example. It makes a lot of sense to use Intel's N150 chip inside a NAS; it is more than capable of doing what the ZimaBoard 2 is intended for, media streaming and backup. It also looks like the IceWhale Technology staff are quite active in the official forums helping people with issues they come across with ZimaOS and the devices, peer support seems to be good as well, I was quickly able to find why I was not able to create a new Storage Pool in ZimaOS v1.6.1 even though that is quite a serious bug, hopefully it will be fixed in the next update. If you are comfortable with the command line and Docker, you'll be fine. You can do great things with this hardware. This was my first time with ZimaOS. It seems a bit barebones in comparison to the likes of Synology DSM, TOS, and UGOS, but it has a ton of apps to get you started with your home or small business NAS. Where to buy As of publishing, IceWhale Technology is running a discount of up to 5% for the Starter Kit. If you opt to get just the ZimaBoard 2 itself, it does come with a SATA Y-Cable, so you will be able to connect up to two 3.5-inch HDDs to it. ZimaBoard 2 1668 Starter Kit for $534.50 on Amazon US (was $548.60) ZimaBoard 2 832 Starter Kit for $372.88 on Amazon US (was $390.60) Zimaboard 2 1668 (16GB+64GB) for $419.90 on Amazon US Zimaboard 2 832 (8GB+32GB) for $359.90 on Amazon Disclosure: IceWhale Technology provided a free sample without any editorial input or review pre-approval. Good to know The Amazon link is U.S. specific, and not available in other regions unless specified. We only use first-party seller links (at the time of article publishing); ensure that you purchase from a first-party seller link only. Check out Today's Deals on Amazon | or our recent tech deals. Become a Prime member (for Students or SNAP) via Neowin Get Prime Access - Prime for half price (for qualifying Medicaid, EBT, SNAP) Subscribe to Prime Video, Audible Plus, Music Unlimited or Kindle Unlimited via Neowin As an Amazon Associate, when you purchase through links on our site, we earn from qualifying purchases.
    • It's in the Insider's group so yes it's technically beta, though these days it's hard to see much of a difference unless you opt for the most extreme beta builds, which I don't. When I moved here from the Release Preview channel I did so primarily because I wanted to see how well the restored taskbar functionality (restored from Win10, and earlier) is working and whether it was time to finally abandon SAB--and it is--working fine, so far. Not as polished as SAB, but it'll do for me.
    • I've been using MWB Premium for a number of years so that along with Windows updates and updated browser should be fine. Thanks for that.
  • Recent Achievements

    • Week One Done
      flexorcist earned a badge
      Week One Done
    • One Month Later
      Woland13 earned a badge
      One Month Later
    • Week One Done
      Woland13 earned a badge
      Week One Done
    • One Year In
      bernmeister earned a badge
      One Year In
    • Week One Done
      Scoobystu earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      488
    2. 2
      +Edouard
      220
    3. 3
      PsYcHoKiLLa
      147
    4. 4
      Steven P.
      74
    5. 5
      FloatingFatMan
      70
  • Tell a friend

    Love Neowin? Tell a friend!