Home Blog Page 2

Check on a fleet of bodily gadgets with Android Gadget Streaming, now with Android Companion Gadget Labs



Check on a fleet of bodily gadgets with Android Gadget Streaming, now with Android Companion Gadget Labs

Posted by Adarsh Fernando – Group Product Supervisor, Android Studio, and Grant Yang – Sr. Product Supervisor, Omnilab

Immediately, we’re excited to provide you an replace on Android Gadget Streaming and announce that Android Companion Gadget Labs are actually steady and accessible within the newest steady launch of Android Studio Narwhal Characteristic Drop!

moving image of Android Device Streaming functionality in Android Studio Narwhal Feature Drop

Streamline your testing with Android Gadget Streaming

Android Gadget Streaming, powered by Firebase, permits you to securely hook up with distant bodily Android gadgets hosted in Google’s safe information facilities. This implies you’ll be able to take a look at your app on all kinds of gadgets and Android variations with out leaving Android Studio, or having to buy each gadget you need to take a look at. This helps you:

    • Check on the newest {hardware}: Get your palms on the most recent gadgets, together with these not but accessible to the general public.
    • Cowl a variety of gadgets: Check on varied type elements, from telephones to foldables, from a large number of producers.
    • Enhance developer productiveness: Save time and assets by testing on actual gadgets immediately out of your growth atmosphere.

At Google I/O 2025 earlier this 12 months, we introduced that Android Gadget Streaming graduated to steady. With that, we added the newest Google Pixel gadgets to the catalog, together with the Pixel 9, Pixel 9 Professional, Pixel 9 Professional XL, and Pixel 9 Professional Fold. We’re additionally working so as to add the upcoming Pixel 10 gadgets to the catalog quickly.

We additionally introduced that we’re partnering with main OEMs to carry you an excellent better collection of gadgets by means of Android Companion Gadget Labs. And better of all, with our month-to-month quota of minutes, you’ll be able to begin testing Android Gadget Streaming with a variety of gadgets for gratis. Utilization past the month-to-month quota of minutes could incur a cost, as described on the Firebase pricing web page.

Introducing Android Companion Gadget Labs

Android Companion Gadget Labs are actually steady and accessible in Android Studio Narwhal Characteristic Drop. Android Companion Gadget Labs provide you with entry to a fleet of bodily gadgets from our OEM companions, together with Samsung, Xiaomi, OPPO, OnePlus, vivo, and extra. This provides you the power to check your apps on the particular gadgets your customers have, making certain a greater consumer expertise for everybody.

Listed below are a few of the gadgets accessible as we speak by means of our companions:

Android Partner Device Labs featuring a fleet of physical devices from our OEM partners, including OPPO, OnePlus, Samsung, Xiaomi, and vivo, in Android Studio Narwhal Feature Drop

Get began with associate gadgets

To begin utilizing gadgets from Android Companion Gadget Labs, comply with these steps in Android Studio:

  1. Open the Gadget Supervisor by navigating to View > Device Home windows > Gadget Supervisor.
  2. Click on on the Firebase icon and log in to your Google Developer account.
  3. Choose a Firebase challenge with billing enabled.
  4. You’ll now see the listing of accessible gadgets, together with these from our companions.

For staff growth, a Firebase challenge administrator (Proprietor or Editor) might want to allow entry to every OEM’s lab of gadgets. To take action, the admin ought to go to the Google Cloud challenge web page, guarantee the right challenge is chosen, after which allow the specified gadget lab by utilizing the toggle and following the on-screen prompts. As soon as enabled, your entire staff may have entry to these gadgets in Android Studio.

Android Gadget Streaming Pricing

You may be taught extra about Android Gadget Streaming quota and pricing on the Firebase pricing web page. Gadgets within the Android Companion Gadget Labs can be found on the identical pricing and with the identical month-to-month quota of minutes for gratis as all different gadgets within the Android Gadget Streaming catalog (except in any other case specified).

Wrapping it up

Android Gadget Streaming, now with the addition of Android Companion Gadget Labs, provides you an unparalleled collection of bodily gadgets to check your apps. With a rising catalog of gadgets from Google and our OEM gadget companions, you’ll be able to guarantee your app works flawlessly for all of your customers.

We invite you to obtain the newest steady launch of Android Studio and check out Android Gadget Streaming and the brand new Android Companion Gadget Labs. With our beneficiant month-to-month quota, you can begin testing on a variety of gadgets for gratis. We’re continuously updating the gadget catalog, so make sure to test again usually for brand new additions.

Completely satisfied streaming!

android – Google Signal-In Failed with “Entry blocked: Authorization error, Error: Invalid_scope” in Flutter


I am getting an “Entry blocked: Authorization error, Error: Invalid_scope” error when attempting to implement Google Signal-In in my Flutter utility with a number of OAuth scopes.

Error Message

Entry blocked: Authorization error, Error: Invalid_scope

My Setup

Flutter Model: [Your Flutter version]
google_sign_in package deal: [Your package version]
Platform: Android
Goal: Google Calendar, and fundamental profile entry

OAuth Request Particulars
My OAuth request consists of the next parameters:

login_hint=ak**********@gmail.com  
hl=en-IN  
response_type=consent_proof  
android_options=CAMQAQ  
color_scheme=darkish  
client_id=*200-m552****************n15f59*p7****kr*u.apps.googleusercontent.com  
include_granted_scopes=true  
scope=https://www.googleapis.com/auth/userinfo.e-mail   https://www.googleapis.com/auth/calendar https://www.googleapis.com/auth/hold openid   https://www.googleapis.com/auth/userinfo.profile  
flowName=GeneralOAuthFlow

Google Cloud Console Configuration
✅ Created OAuth 2.0 Shopper ID for Android
✅ Added appropriate package deal identify and SHA-1 fingerprint
✅ Enabled Google Calendar API
✅ Enabled Google Maintain API (if accessible)
✅ Added check customers to OAuth consent display screen

Code Implementation

import 'package deal:google_sign_in/google_sign_in.dart';

class GoogleSignInService {
  static remaining GoogleSignIn _googleSignIn = GoogleSignIn(
    scopes: [
      'email',
      'profile',
      'openid',
      'https://www.googleapis.com/auth/calendar',
      'https://www.googleapis.com/auth/keep',
    ],
  );

  static Future signIn() async {
    attempt {
      remaining GoogleSignInAccount? account = await _googleSignIn.signIn();
      return account;
    } catch (error) {
      print('Google Signal-In Error: $error');
      return null;
    }
  }
}

home windows – TCP Packets Despatched from Shopper By no means Arrive at Server


I am dealing with a networking subject with my MMORPG server setup. The server infrastructure is as follows:

Host: Home windows Server
Server Software program and Companies: Hosted inside a Digital Machine operating Debian
Networking: The VM is bridged/NATed (specify as wanted)

For some accounts, sure packets (e.g., packet 0x09a1, 2 bytes in size) are despatched from the consumer (confirmed by native Wireshark and an exterior library hooked into the consumer), however by no means arrive on the server—Wireshark operating on the Debian VM by no means sees them. For different accounts, the whole lot works tremendous and packets are obtained as anticipated.

Technical Particulars

The problematic packets are clearly seen as despatched in native Wireshark captures.

On the server (Debian VM), those self same packets are fully lacking from the seize when utilizing a problematic account.

When evaluating native vs server captures, I generally discover 4 additional bytes added on the finish of the packet (e.g., f7 38 dc d7), for working accounts.

If I modify to a working account ID, the packets are obtained appropriately, and the additional 4 bytes aren't current.

On the applying facet, the related breakpoint is rarely hit, confirming the packet by no means reaches the server logic.

Different TCP site visitors between consumer and server works usually. Solely sure packets with particular account IDs "disappear."

No particular firewall guidelines or community restrictions are configured for particular accounts or packets.

Extra Evaluation

Cloned VM Works: I made an an identical copy of the Debian VM (identical configuration, identical server software program, identical knowledge) and ran it domestically on my machine. On this setup, the whole lot works completely—even for the problematic accounts, each domestically and for exterior customers connecting from exterior.

SSH Forwarding Works: I additionally examined connecting by way of SSH port forwarding to the stay server. When forwarding the emulator ports by way of SSH, the connection works tremendous for these accounts—the lacking packets arrive as they need to.

Suspect: The difficulty appears remoted to the networking path between the consumer and the Debian VM when operating on the manufacturing Home windows Server host. No OS/software logs report dropped packets or errors.

Has anybody skilled a state of affairs the place TCP packets despatched from the consumer by no means arrive on the server, however just for particular account IDs—and solely on a specific Home windows Server + Debian VM setup? Might this be attributable to deep packet inspection, host firewall, virtualization networking settings, or sure offloading options (TCP Offloading, checksum offload, and many others.) on the Home windows Server host?

Any solutions or diagnostic steps can be extremely appreciated!

Android Builders Weblog: Accelerating growth with month-to-month releases for Android Studio



Android Builders Weblog: Accelerating growth with month-to-month releases for Android Studio

Posted by Xavier Ducrohet – Tech Lead, Android Studio and Adarsh Fernando – Group Product Supervisor, Android Studio

Final yr, we doubled our launch frequency for Android Studio with the introduction of Function Drops, a change designed to deliver you new options and enhancements extra shortly. Immediately, we’re excited to announce the subsequent evolution in our launch schedule: we’re transferring to month-to-month secure releases of Android Studio.

This new cadence means you’ll be capable of get your palms on the newest options and important enhancements, quicker than ever earlier than. This is what you may anticipate: each few months, we’ll introduce a model that accommodates the newest IntelliJ platform model, comparable to Android Studio Narwhal primarily based on IntelliJ 2025.1. You’ll then see Function Drops every month that embody vital bug fixes and new performance you’ll need to check out, till it’s time to launch Android Studio with the subsequent platform model of IntelliJ.

Android Studio Narwhal release cadence 2025

You’ve really already skilled this new launch cadence with Android Studio Narwhal! Within the final Function Drop launch we had been in a position to take options comparable to Agent Mode from Canary to the Steady channel quicker than ever earlier than, making it doable so that you can check out new options, quicker!

Why Month-to-month Releases?

You advised us ready for the subsequent main launch to get a vital bug repair or a quality-of-life enchancment might be irritating. With the transfer to month-to-month releases, we will ship these updates to you with out the lengthy delays. This implies you may have entry to the options you need and the fixes you want, proper once you want them.
It is vital to notice that the Android Emulator and the Android Gradle Plugin will proceed to be up to date individually from Android Studio at a tempo of each two months. And, as at all times, you need not replace these parts to obtain and use the newest secure model of the IDE every month.

Our Dedication to High quality

A quicker launch cadence does not imply a compromise on high quality. In reality, our potential to launch extra continuously is a direct results of our long-term funding in our testing infrastructure. This effort started with Venture Marble, a concerted effort to enhance the standard and testing of Android Studio. Since then, we have been repeatedly enhancing and tooling our testing technique to be extra dependable, and to get suggestions from these checks quicker. Final yr, we reached some extent the place we may confidently double our releases, and now, we’re able to take the subsequent step with month-to-month updates. This implies you’ll see releases 2X extra usually than earlier than!

We’re additionally persevering with to supply early entry to stable-ready releases. Beforehand, we’ve supplied these alternatives first by way of Beta releases. With our investments in detecting and addressing points earlier, we’re in a position to take a launch from Canary immediately into our Launch Candidate (RC) channel with a stable-ready degree of high quality and polish. This provides you an opportunity to check out the newest options and enhancements earlier than they’re launched to the secure channel and to supply us with useful suggestions.

Replace Month-to-month and Assist Us Enhance

We encourage you to replace to the newest secure model of Android Studio every month to reap the benefits of the newest options and enhancements. Your suggestions is important to serving to us make Android Studio one of the best it may be.

This is how one can become involved:

    • Obtain Android Studio Narwhal 3 Function Drop: It’s at present out there within the Canary channel and is the easiest way to get early entry to new options and to supply us with suggestions earlier than a launch is finalized. If you need a extra secure construct, obtain this model as quickly because the Launch Candidate turns into out there.
    • Report a bug: In case you encounter a problem, please tell us by reporting a bug. This helps us to determine and repair points extra shortly.

We’re enthusiastic about this new chapter for Android Studio and we’re assured that it’s going to aid you to construct higher apps, quicker. As at all times, you might be a part of our vibrant Android developer neighborhood on LinkedIn, Medium, YouTube, or X.



AI finds extra breast tumors sooner than conventional double radiologist evaluate – NanoApps Medical – Official web site


AI is detecting tumors extra usually and earlier within the Dutch breast most cancers screening program. These tumors can then be handled at an earlier stage. This has been demonstrated by researchers led by Radboud college medical middle in a research revealed in The Lancet Digital Well being. The usage of AI may scale back workload and save thousands and thousands of euros yearly.

Earlier analysis in Sweden had already proven that AI detects breast most cancers on mammograms extra steadily than radiologists. Furthermore, AI can scale back the workload for radiologists. Now, it seems that AI may change the second radiologist within the Dutch breast most cancers screening program. This even results in the detection of extra tumors—and at an earlier stage—which later transform clinically vital.

AI is extra prone to be proper

Researchers, led by breast radiologist Ritse Mann of Radboudumc, analyzed 42,000 breast scans. These mammograms have been taken as a part of the Dutch screening program within the Utrecht area.

Historically, two radiologists evaluate these scans, as is normal follow in breast most cancers screening. On this research, the researchers additionally evaluated the scans utilizing AI developed by ScreenPoint Medical. Moreover, they adopted the ladies whose scans have been analyzed for practically 4 and a half years, with a number of scans accessible for a lot of of them.

The research confirmed that one radiologist working with AI detects extra tumors than two radiologists alone. Tumors are additionally recognized earlier when AI is concerned.

“Generally the AI detects a tumor that the radiologists don’t but acknowledge as such. We name this a false optimistic. However usually that tumor seems in a later scan in any case. Due to this fact, the AI was proper earlier,” Ph.D. candidate Suzanne van Winkel explains.

“By the point the radiologist raises the alarm, it usually considerations bigger invasive tumors, which undoubtedly want therapy, as early as attainable.”

Changing the second radiologist

In Sweden, AI is already getting used to research screening mammograms. “They change the second radiologist with AI. Provided that the AI is unsure does a second radiologist step in,” Mann explains.

“We see that radiologists work nicely with AI, which ends up in extra tumors being detected and not using a vital enhance in pointless follow-up checks for ladies.”

The present research reveals that AI can be adequate within the Dutch setting to interchange the second radiologist in screening scan evaluation. This might save a number of million euros per yr. But AI is just not at present getting used within the Netherlands.

“In Sweden, screening is organized regionally, whereas within the Netherlands it’s organized nationally. That makes implementing AI extra logistically difficult right here,” says Mann. “The IT infrastructure within the Netherlands is just not but prepared. That requires funding.”

Extra info: AI detects further clinically related breast cancers as an impartial second reader inside a population-based screening program: a retrospective research, The Lancet Digital Well being (2025).

Supplied by Radboud College